JP3014328B2 - Data search method and link information addition method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data search method in a data structure that can constitute, for example, a database or artificial intelligence, and a link information adding method for adding link information to a data string of such a data structure.

[0002]

2. Description of the Related Art At present, various databases are put into practical use. Among them, the most widely used database is a relational database. Various data are accumulated in the database, and the data accumulated in the database is searched using various methods.

For example, concept names such as "desk", "pencil", "tree", "graphite", "mica" and "motor brush" are composed of respective element names as shown in Table 1 below. An outline of a conventional relational database will be described below, taking as an example a case where a concept name is recorded in a concept field and each element name is recorded in an element field in each record. In Table 1, numbers indicate record numbers, and alphabets indicate element field numbers.

[0004]

[Table 1]

In relational databases, X ｛..., Y,.
A new record relationship can be created by deriving the relationship of X {Z} from the combination of the two records of. The uppercase alphabet before the curly bracket "@" indicates the concept name, and the uppercase alphabet within the curly bracket "@" indicates the element name. For example, ｛,｝ and ｛, flammable,
The concept name “desk” and the element name “flammable” can be combined from the combination of the records “｝” as in “flammable”. For convenience, the record "X" indicates a record in which the concept name "X" is recorded in the concept field.

[0006]

In a relational database, for example, the concept name "desk" is replaced by "flammable".
In order to derive the information that the element name can be included, that is, to combine the record “X” including the element name “Y” and the record “Y” including the element name “Z”, the element name “ It is necessary to know what element field of record "X" contains "Y". For example, in the example shown in Table 1, it is necessary to know that the element name "tree" is recorded in the element field number A of the record number 1. However, it is not always known which element field contains the element name “Y.” Therefore, all the element fields of record X are combined with the concept fields of other records. However, in such an operation,
When the number of records and the number of element fields increase, enormous processing is required. In order to prevent such a situation, there is a method in which one record is composed of one concept field and one element field, and the element field of the record is combined with the concept field of another record. Not only is it enormous, but you also have to enter the same concept name in the concept field of many records.

In the relational database, if the element name "flammable" is searched without combining the records, the record "pencil" and the record "tree" are extracted. That is, it is possible to extract the concept name directly related. However, the record “desk” does not include the element name “flammable” in the element field, although the element name “tree” may be included in the record “desk”. No information can be derived that it can. That is, it is not possible to indirectly extract related concept names. Therefore, in the relational database, in order to solve this problem, the element name “flammable” must be entered in the element field of the record “desk”. That is, there is a problem that the same element name must be repeatedly input to the element fields of a plurality of records. In addition, many element names must be entered in the element field of a record having a certain concept name. Or, alternatively,
It is sufficient to combine all element fields with concept fields, but such a combination creates huge data and it is unknown which element field contains the element name to be searched. You need to search element fields.

Furthermore, in relational databases, it is not possible to prevent the occurrence of cyclic definitions such as A {B}, B {C}, and C {A}. If the existence of such a circularly defined record is permitted, an infinite loop occurs when the search is performed.

Further, in general, in a relational database, original data is arranged in the form of a two-dimensional table. Each column corresponds to a concept field and an element field containing fixed and similar data, the data being pre-sorted and entered into each field. Therefore, it includes an infinite number of element names like the concept name "pencil",
Moreover, a relational database is not suitable for expressing a case where the attributes of data to be input to the element fields of each element name are not classified.

In the relational database described above, the concept fields are stored in the same table. However, the same applies to the case where one table is created from one concept.

Therefore, a first object of the present invention is to provide a data search method capable of extracting not only directly related concept names but also indirectly related concept names. Further, a second object of the present invention is to provide a link which can eliminate the occurrence of the above-described cyclic definition when adding new link information to a predetermined area of a data string in a data string aggregate composed of a plurality of data strings. It is to provide a method for adding information.

[0012]

According to a first aspect of the present invention, there is provided a data search method for a data identifier corresponding to a parent concept for a data identifier to be searched. The present invention relates to a data search method for extracting a column and / or a data column corresponding to an indirect parent concept. The parent concept and the indirect parent concept will be described later in detail. That is, in the data search method according to the first aspect of the present invention, each data string (DS_1) includes (a) a data identifier area in which a data identifier is recorded, and (b) a data level value. (C) a data level value area, and (c) a parent data area in which link information on a data string (DS_2) in which a data identifier indicating a set including the data identifier is recorded is recorded. The data level value of (DS_2) is larger than the data level value of the data sequence (DS_1). And all the link information recorded in the parent data area of the data string (DS_1) are different. (C) Starting from any link information recorded in the parent data area of the data string (DS_1) When the data string is sequentially traced based on the link information recorded in the parent data area of the data string until the parent data area of the data string becomes empty, the link information relating to the data string (DS_1) (A) inputting a data identifier corresponding to data to be searched for in a data string aggregate composed of a plurality of data strings that is not included in a parent data area of the data string of (a). And (b) extracting the data string based on the link information recorded in the parent data area of the data string whose input data identification name is recorded in the data identification name area, and adding the extracted data string to the parent data area of the extracted data string. A step of extracting a data string based on the recorded link information until a parent data area of the extracted data string becomes empty; and (c) outputting the extracted data string. The features.

Here, the link information is a direct address or an indirect address indicating an area where a data string is recorded,
A pointer indicating an address area in which a data string is recorded, or a data identification name or the like recorded in the data string. If the information can be reached,
Any information may be used. The same applies to link information in the following description.

In the data search method according to the first aspect of the present invention, as schematically shown in FIG. 1, first, in step (b), the input data identification name is recorded in the data identification name area. Data strings P ₁ , P ₂ , P ₃ ,... Based on the link information recorded in the parent data area of the data string S _0.
・ Extract Then, the data strings P ₁ , P ₂ , P ₃ ,.
.. Regarding these extracted data strings P ₁ , P ₂ ,
Data strings are extracted based on the link information recorded in the parent data area of P ₃ ,. The extracted data column is
For example, data strings P ₁₁ , P ₁₂ , P ₁₃ ..., P ₂₁ , P ₂₂ ,
_{P 23 ···, P 31, P} 32, and P ₃₃ · · · (the (A) see Figure 1). Note that FIG. 1 shows only a part of the data string.

Then, all the extracted data strings P ₁₁ , P ₁₂ , P ₁₃ ..., P ₂₁ , P ₂₂ , P ₂₃ .
Data strings are extracted based on the link information recorded in the parent data areas of ₃₁ , ₃₂ , _33, ... The extracted data strings are P ₁₁₁ , P ₁₁₂ , P ₁₁₃ ..., P ₁₂₁ , P ₁₂₂ , P
₁₂₃ ..., _P131 , _P132 , _P133 ..., _P211 ,
_{P 2 12, P 213 ···,} P 221, P 222, P 223 ···, P
₂₃₁ , P ₂₃₂ , P ₂₃₃ ..., P ₃₁₁ , P ₃₁₂ , P ₃₁₃ ...
_{·, P 321, P 322,} P 323 ···, P 331, P 332, P 33 3
(See FIG. 1B). Next, a data string is extracted based on the link information recorded in the parent data area of these data strings P ₁₁₁ , P ₁₁₂ , P ₁₁₃ . Such processing is repeated until the parent data area of the extracted data string becomes empty. The above processing flow is referred to as search pattern 1 for convenience. In FIG. 1, the symbols in the ellipse represent data strings, and the numbers attached to the straight lines connecting the ellipse and the ellipse represent the order of processing.

The above-mentioned step (b) can be performed according to the processing flow (referred to as search pattern 2) shown in FIG. That is, In the step (b), first, extracts the data string P ₁ based on the link information, the input data DN is recorded in the parent data region of the data sequence S ₀ recorded in the data identification name area. FIG. 2 shows only a part of the data string.

[0017] Then, based on the extracted data sequence link information recorded in the parent data region of P _1, the parent data area of the extracted data sequence until empty, to extract the data string. That is, the data sequence P ₁₁ , the data sequence P ₁₁₁ , the data sequence P
₁₁₁₁ is sequentially extracted. Since the parent data area of the extracted data sequence P ₁₁₁₁ is empty, return to the data string P _111, based on the data string P _{1 11} next link information recorded in the parent data region of, extracts a data sequence P _1112. Since the parent data region of the extracted data string P ₁₁₁₂ is empty, return to the data string P _111, based on the next link information recorded in the parent data area of a data string P _111, extracts a data sequence P _1113. Since the parent data region of the extracted data string P ₁₁₁₃ is empty, the process returns to the data string P _111, since the next parent data area of a data string P ₁₁₁ is empty, return to the data string P _11, the data string P ₁₁ following , The data string P ₁₁₂₁ , the data string P ₁₁₂₂ , the data string P ₁₁₂ based on the link information recorded in the parent data area of
₁₁₂₃ is sequentially extracted. Such processing is sequentially performed. Then, if the link information recorded in the parent data region of the data sequence P ₁₁ is empty, based on the link information recorded in the next parent data area of a data string P _1, extracts a data sequence. Furthermore, if the link information recorded in the parent data area of a data string P ₁ is empty, on the basis of the data string next link information recorded in the parent data area of S _0, the data string P _2, P ₃ , ... are extracted. In FIG. 2 as well, the symbols in the ellipse represent data strings, and the numbers attached to the straight lines connecting the ellipse and the ellipse represent the order of processing.

According to a second aspect of the present invention, there is provided a data retrieval method for achieving the first object, wherein a data sequence corresponding to an element concept and / or
Alternatively, the present invention relates to a data search method for extracting a data string corresponding to an indirect element concept. The element concept and the indirect element concept will be described later in detail. That is, in the data search method according to the second aspect of the present invention, each data string (DS
_1) are (a) a data identifier name area in which a data identifier is recorded, (b) a data level value area in which a data level is recorded, and (c) an element of a set represented by the data identifier. And (A) a data level value of the data sequence (DS_0) is higher than a data level value of the data sequence (DS_1). (B) The element data area of the data string (DS_1) does not include link information relating to the data string (DS_1) itself, and the link information recorded in the element data area of the data string (DS_1) is (C) Starting from arbitrary link information recorded in the element data area of the data string (DS_1), the data string is required until the element data area of the data string becomes empty. When the data string is sequentially traced based on the link information recorded in the data area, the link information on the data string (DS_1) is obtained from a plurality of data strings that are not included in the element data area of any traced data string. A data search method for a structured data string aggregate, comprising: (a) inputting a data identifier corresponding to data to be searched; and (b) recording the input data identifier in a data identifier area. The data string is extracted based on the link information recorded in the element data area of the extracted data string, and the element data area of the extracted data string is extracted based on the link information recorded in the element data area of the extracted data string. Extracting the data sequence until it is empty; (c) outputting the extracted data sequence;
Characterized by comprising:

In the data search method according to the second aspect of the present invention, as schematically shown in FIG. 3, first, in step (b), the input data identification name is recorded in the data identification name area. Based on the link information recorded in the element data area of the data string S ₀ , the data strings E ₁ , E ₂ ,
E ₃ , ... are extracted. Then, the data strings E ₁ , E ₂ ,
With respect to E ₃ ,..., These data strings E ₁ , E ₂ ,
Data strings are extracted based on the link information recorded in the element data areas of E ₃ ,. The extracted data strings are, for example, data strings E ₁₁ , E ₁₂ , E ₁₃ ..., E ₂₁ , E
₂₂ , E ₂₃ ..., E ₃₁ , E ₃₂ , E ₃₃ .
(A)). FIG. 3 shows only a part of the data string.

Next, all the extracted data strings E ₁₁ , E ₁₂ , E ₁₃ ..., E ₂₁ , E ₂₂ , E ₂₃ .
_31, extracts the data sequence on the basis of the E _32, E ₃₃ ··· element data area recorded link information to the. The extracted data _{sequence, E 111, E 112, E} 11 3 ···, E 121, E 122,
_E123 ..., _E131 , _E132 , _E133 ..., _E211 , E
₂₁₂ , _E213 ..., _E221 , _E222 , _E223 ...,
E ₂₃₁ , E ₂₃₂ , E ₂₃₃ ..., E ₃₁₁ , E ₃₁₂ , E _{313.
·, E 321, E 322,} E 323 ···, E 331, E 332, E 33 3
(See FIG. 3B). Next, data strings are extracted based on the link information recorded in the element data areas of these data strings E ₁₁₁ , E ₁₁₂ , E ₁₁₃ . Such processing is repeated until the element data area of the extracted data string becomes empty. The above processing flow is referred to as a search pattern 3 for convenience. In FIG. 3, the symbols in the ellipse represent data strings, and the numbers attached to the straight lines connecting the ellipse and the ellipse represent the order of processing.

The above-mentioned step (b) can be performed according to the processing flow (referred to as a search pattern 4) shown in FIG. That is, In the step (b), first, on the basis of the link information input data DN is recorded in the element data area of the data sequence S _0, which is recorded in the data DN region, extracts a data sequence E ₁ . In FIG. 4, the symbols in the ellipse represent data strings, and the numbers attached to the straight lines connecting the ellipse and the ellipse represent the order of processing. Then, based on the recorded link information extracted element data area of a data string E ₁ was, element data area of the extracted data sequence until empty, to extract the data string. That is, the data sequence E ₁₁ , the data sequence E ₁₁₁ , and the data sequence E ₁₁₁₁ are sequentially extracted. Since element data area of the extracted data sequence E ₁₁₁₁ is empty, return to the data sequence E _111, based on the data string next element data area recorded link information of the E _111, extracts a data sequence E _1112. Since element data area of the extracted data sequence E ₁₁₁₂ is empty, return to the data sequence E _111, based on the data string next element data area recorded link information of the E _111, extracts a data sequence E _1113. Since the element data area of the extracted data string E _{11 13} is empty, the data string E ₁₁₁
Returning to, the following elements data area of the data sequence E ₁₁₁ is empty, return to the data sequence E _11, based on the data string next element data area recorded link information of the E _11, a data sequence E
₁₁₂ , data string E ₁₁₂₁ , data string E ₁₁₂₂ , data string E
Extract ₁₁₂₃ . Such processing is sequentially performed. Then, if the link information recorded in the element data area of the data sequence E ₁₁ is empty, based on the link information recorded in the next element data area of a data string E _1, extracts a data sequence. Furthermore, if the link is recorded in the element data area of a data string E ₁ information becomes empty, based on the data string next element data area recorded link information of S _0,
Data strings E ₂ , E ₃ ,... Are extracted.

According to a third aspect of the present invention, there is provided a data search method for achieving the first object, wherein a data sequence corresponding to a parent concept and a indirect parent concept are added to a data identifier to be searched. The present invention relates to a data search method for extracting at least one of a corresponding data string, a data string corresponding to an element concept, and a data string corresponding to an indirect element concept. That is, in the data search method according to the third aspect of the present invention, each data string (DS_1) includes (a) a data identification name area in which a data identification name is recorded, and (b) a data level value. A data level value area, (c) a parent data area in which link information relating to a data string (DS_2) in which a data identifier indicating a set including the data identifier is recorded, and (d) a data identifier represented by the data identifier And (A) an element data area in which link information relating to a data string (DS_0) in which a data identifier that is an element of a set to be recorded is recorded; (A) a data level value of the data string (DS_2) is (B) The parent data area of the data string (DS_1) does not include link information about the data string (DS_1) itself, and the data string (D_1) _1), the link information recorded in the parent data area is different, and (C) the parent data area of the data string becomes empty starting from any link information recorded in the parent data area of the data string (DS_1). Up to this point, when the data string is sequentially traced based on the link information recorded in the parent data area of the data string, the link information relating to the data string (DS_1) is not included in the parent data area of any traced data string. (D) The data level value of the data string (DS_0) is smaller than the data level value of the data string (DS_1). (E) The element data area of the data string (DS_1) relates to the data string (DS_1) itself. Link information that does not include link information and is recorded in the element data area of the data string (DS_1) is all different. (F) Element data area of the data string (DS_1) When the data sequence is sequentially traced based on the link information recorded in the element data area of the data string until the element data area of the data string becomes empty starting from any recorded link information, the data string (DS_1) The link information related to the data search method is a data search method in a data string aggregate composed of a plurality of data strings, which is not included in the element data area of any of the traced data strings, Inputting a corresponding data identifier, and (b) extracting and extracting the data string based on the link information recorded in the parent data area of the data string whose input data identifier is recorded in the data identifier area. Based on the link information recorded in the parent data area of the extracted data string, the data string is extracted until the parent data area of the extracted data string becomes empty. A data string is extracted based on the link information recorded in the element data area of the data string recorded in the data identification name area, and the link information recorded in the element data area of the extracted data string is Based on
The method comprises the steps of: extracting a data string until the element data area of the extracted data string becomes empty; and (d) outputting the extracted data string.

The flow of the search processing in the data search method according to the third aspect of the present invention includes the search pattern 1 or search pattern 2 described in the data search method according to the first aspect of the present invention, Since the search pattern 3 or the search pattern 4 described in the data search method according to the second aspect may be appropriately combined, a detailed description thereof will be omitted.

According to a fourth aspect of the present invention, there is provided a data search method for achieving the above first object, wherein a data string corresponding to an element concept and / or
Or, a data string corresponding to the indirect element concept and extracting a data string having the lowest data level value, and then, for the data identifier recorded in the extracted data string,
The present invention relates to a data search method for extracting a data sequence corresponding to a parent concept and a data sequence corresponding to an indirect parent concept. That is, the data search method according to the fourth aspect of the present invention comprises: (a) inputting a data identifier corresponding to data to be searched;
(B) The data string to be searched is extracted based on the link information recorded in the element data area of the data string recorded in the data identifier name area, and is recorded in the element data area of the extracted data string. Extracting the data string until the element data area of the extracted data string becomes empty based on the link information, and (c) setting the element data area in the data string extracted in the step (b). The data string is extracted based on the link information recorded in the parent data area of the empty data string, and the parent data area of the extracted data string is extracted based on the link information recorded in the parent data area of the extracted data string. It is characterized by comprising a step of extracting a data string until it becomes empty, and a step (d) of outputting the extracted data string. The structure of the data string aggregate in the data search method according to the fourth aspect of the present invention may be the same as the structure of the data string aggregate in the data search method according to the third aspect of the present invention. it can.

In the data search method according to the fourth aspect of the present invention, the flow of the search processing in the step (b) is the same as the search pattern 3 or the search pattern described in the data search method according to the second aspect of the present invention. Pattern 4 may be used. On the other hand, the flow of the search process in step (c) may be search pattern 1 or search pattern 2 described in the data search method according to the first embodiment of the present invention.
Detailed description is omitted.

According to the first aspect of the present invention, there is provided a link information adding method for achieving the second object, wherein a parent identification number area constituting a parent data area of a data string (DS_I) is provided. And a link information adding method for adding new link information relating to a data string (DS_P) corresponding to a parent concept.
That is, in the link information adding method according to the first aspect of the present invention, each data string (DS_1) includes (a) a data identification name area in which a data identification name is recorded, and (b) a data level value. (C) a parent identification number area for recording link information relating to a data string (DS_2) recording a data identification name indicating a set including the data identification name , the data string (DS_2)
The parent data area, which has or more than the number of
Consists, data level values of (A) data strings (DS_2) is greater than the data level values of the data sequence (DS_1), constituting the parent data area (B) data strings (DS_1)
The data string (DS_1) is stored in any of the parent identification number areas.
The link information that does not include the link information relating to itself and that is recorded in the parent identification number area constituting the parent data area of the data string (DS_1) is all different. (C) The parent of the data string (DS_1) Configure the data area
Starting from the link information recorded in an arbitrary parent identification number area, the parent identification number area constituting the parent data area of the data string
Until the area becomes empty, the parent that constitutes the parent data area of the data string
When the data sequence is sequentially traced based on the link information recorded in the identification number area , the link information on the data sequence (DS_1) indicates the parent data area of any traced data sequence.
A link information adding method for adding new link information to a parent identification number area constituting a parent data area of a data string in a data string aggregate composed of a plurality of data strings which is not included in the constituting parent identification number area (A) inputting a data identification name and inputting a parent concept name indicating a set including the data identification name; and (b) a parent inputting the input data identification name. If it matches the concept name, stop adding link information. (C) If the input data identification name does not match the input parent concept name, a data string (DS_I) recording the input data identification name (D) extracting a data string (DS_P) in which a data identifier matching the input parent concept name is recorded; and (e) extracting the data string extracted in step (c). (DS_I)
If link information relating to the data string (DS_P) extracted in step (d) exists in the parent identification number area constituting the parent data area of step (c), the addition of link information is stopped, and (f) step (c) (DS_I)
If there is no link information on the data string (DS_P) extracted in the step (d) in the parent identification number area that constitutes the parent data area of the data string (DS_P) extracted in the step (d), Parent identification of parent data area
A data string is extracted based on the link information recorded in the number area , and a parent constituting a parent data area of the extracted data string is extracted.
Based on the recorded link information to the identification number field, a parent ID number area <br/> constituting the parent data area of the extracted data sequence until empty, the step of extracting the data string, (g) step If any of the data strings extracted in (f) match the data string (DS_I) extracted in step (c), the addition of link information is stopped, and (h) the step (f) If none of the data strings extracted in step (c) match the data string (DS_I) extracted in step (c), the number of parent identification number areas becomes insufficient.
If there is, create a new parent identification number area,
A step of recording link information on the data string (DS_P) extracted in the step (d) in a parent identification number area constituting a parent data area of the data string (DS_I) extracted in the step (c); (I) Data string (DS_P) extracted in step (d)
Is less than or equal to the data level value of the data string (DS_I) extracted in the step (c), the data extracted in the step (c) is stored in the data level value area of the data string (DS_P). Recording a value obtained by adding 1 to the data level value of the column (DS_I).

In the link information adding method according to the first aspect of the present invention, the procedure of extracting the data string in step (f) is the same as the search pattern 1 or the search pattern described in the data searching method according to the first aspect of the present invention. It can be similar to pattern 2.

According to a second aspect of the present invention, there is provided a link information adding method for achieving the second object, wherein an element identification number constituting an element data area of a data string (DS_I) is provided .
The present invention relates to a link information adding method for adding new link information relating to a data string (DS_E) corresponding to an element concept to an area . That is, in the link information adding method according to the second aspect of the present invention, each data string (DS_1) includes: (a) a data identifier name area in which a data identifier is recorded; and (b) a data level value. (C) an element identification number area for recording link information relating to a data string (DS_0) recording a data identification name which is an element of a set represented by the data identification name;
As many or more the number of (DS_0), main <br/> raw data area with, is composed of, from the data level value, the data level values of the data sequence (DS_1) of (A) data strings (DS_0) (B) constituting the element data area of the data string (DS_1)
The data string (DS_
1) An element identification number which does not include link information relating to itself and constitutes an element data area of the data string (DS_1)
All the link information recorded in the signal area are different, and (C) constitute the element data area of the data string (DS_1) .
Starting from the link information recorded in the arbitrary element identification number area, the element identifiers constituting the element data area of the data string
Until another number area is empty, the element data area of the data row
When the data sequence is sequentially traced based on the link information recorded in the constituent element identification number area , the data sequence (DS_1)
Links infos are not included in the element identification number field that constitutes the element data area of any data sequence followed, in a data string assembly comprising a plurality of data strings, the element data area of a data string Element identification number to configure
A link information adding method for adding new link information to a data area, comprising: (a) inputting a data identification name and inputting an element concept name which is an element of a set represented by the data identification name; (B) if the input data identifier matches the input element concept name, stop adding link information; and (c) if the input data identifier does not match the input element concept name. Extracting a data string (DS_I) in which the input data identification name is recorded, and (d) extracting a data string (DS_E) in which the data identification name that matches the input element concept name is recorded. (E) Data string (DS_I) extracted in step (c)
If link information relating to the data string (DS_E) extracted in the step (d) exists in the element identification number area constituting the element data area of (c), the addition of the link information is stopped, and (f) step (c) (DS_I)
If there is no link information on the data string (DS_E) extracted in the step (d) in the element identification number area constituting the element data area of (a), the data string (DS_E) extracted in the step (d) Elements that make up the element data area
Extracts a data string based on the link information recorded in the identification number area and composes the element data area of the extracted data string
It is necessary to construct the element data area of the extracted data string based on the link information recorded in the element identification number area
(G) extracting one of the data strings extracted in step (f) until the element identification number area becomes empty; ), The addition of link information is stopped, and (h) none of the data strings extracted in step (f) match the data string (DS_I) extracted in step (c). If the number of element identification number area is insufficient
If it is, create a new element identification number area
In, the element identification number field that constitutes the element data area of the process the extracted data sequence in (c) (DS_I), step of recording the link information about the process the extracted data sequence in (d) (DS_E) And (i) the data string (DS_E) extracted in step (d)
Is greater than or equal to the data level value of the data string (DS_I) extracted in step (c), the data level value recorded in the data string (DS_E) extracted in step (d) is 1 Is recorded in the data level value area of the data string (DS_I) extracted in step (c).

In the link information adding method according to the second aspect of the present invention, the procedure of extracting the data string in step (f) is the same as the search pattern 3 or the search pattern described in the data searching method according to the second aspect of the present invention. It can be the same as the pattern 4.

According to a third aspect of the present invention, there is provided a link information adding method for achieving the second object, wherein a parent identification number area constituting a parent data area of a data string (DS_I) is provided. And a link information adding method for adding new link information relating to a data string (DS_P) corresponding to a parent concept.
That is, in the link information adding method according to the third aspect of the present invention, each data string (DS_1) includes (a) a data identification name area in which a data identification name is recorded, and (b) a data level value. (C) a parent identification number area for recording link information relating to a data string (DS_2) in which a data identifier indicating a set including the data identifier is recorded;
The parent data area, which has or more than the number of
And (d) an element identification number area for recording link information relating to a data string (DS_0) recording a data identification name which is an element of a set represented by the data identification name,
As many or more the number of (DS_0), main <br/> raw data area with, is composed of, from the data level value, the data level values of the data sequence (DS_1) of (A) data strings (DS_2) (B) configures the parent data area of the data string (DS_1)
The data string (DS_1) is stored in any of the parent identification number areas.
The link information that does not include the link information relating to itself and that is recorded in the parent identification number area constituting the parent data area of the data string (DS_1) is all different. (C) The parent of the data string (DS_1) Configure the data area
Starting from the link information recorded in an arbitrary parent identification number area, the parent identification number area constituting the parent data area of the data string
Until the area becomes empty, the parent that constitutes the parent data area of the data string
When the data sequence is sequentially traced based on the link information recorded in the identification number area , the link information on the data sequence (DS_1) indicates the parent data area of any traced data sequence.
Not included in the parent ID number regions constituting the data level values of (D) a data string (DS_0) is smaller than the data level values of the data sequence (DS_1), element data area of the (E) data strings (DS_1) Make up
The data string (DS_
1) An element identification number which does not include link information relating to itself and constitutes an element data area of the data string (DS_1)
All the link information recorded in the signal area is different, and (F) constitutes the element data area of the data string (DS_1) .
Starting from the link information recorded in the arbitrary element identification number area, the element identifiers constituting the element data area of the data string
Until another number area is empty, the element data area of the data row
When the data sequence is sequentially traced based on the link information recorded in the constituent element identification number area , the data sequence (DS_1)
Links infos are not included in the element identification number field that constitutes the element data area of any data sequence followed, in a data string assembly comprising a plurality of data strings, the parent data area of a data string Constituent parent identification number area
A link information adding method for adding new link information to an area , comprising: (a) inputting a data identifier, and inputting a parent concept name indicating a set including the data identifier; (b) If the input data identifier matches the input parent concept name, stop adding link information. (C) If the input data identifier does not match the input parent concept name, (D) extracting a data sequence (DS_I) that records the data identification name that has been recorded, and (d) extracting a data sequence (DS_P) that records a data identification name that matches the input parent concept name; ) Data string (DS_I) extracted in step (c)
If link information relating to the data string (DS_P) extracted in step (d) exists in the parent identification number area constituting the parent data area of step (c), the addition of link information is stopped, and (f) step (c) (DS_I)
If there is no link information relating to the data string extracted in step (d) in the parent identification number area constituting the parent data area of step (d), the data string (D
The data string is extracted based on the link information recorded in the parent ID number area constituting the parent data region of S_P), it is recorded in the parent ID number area <br/> constituting the parent data area of the extracted data sequence Extracting the data string based on the link information until the parent identification number area constituting the parent data area of the extracted data string becomes empty; and (g) extracting the data string in step (f). If any of the above matches the data string (DS_I) extracted in step (c), the addition of link information is stopped, and (h) any of the data strings extracted in step (f) If the number does not match the data string (DS_I) extracted in step (c), the number of parent identification number areas is insufficient.
If there is, create a new parent identification number area,
In the parent identification number area constituting the parent data area of the data string (DS_I) extracted in the step (c), link information on the data string (DS_P) extracted in the step (d) is recorded. (I) the data string (DS_P) extracted in step (d)
Element identification number area that constitutes the element data area in
If the number of element identification number areas is insufficient,
Is a step of recording link information relating to the data string (DS_I) extracted in step (c) after newly generating an element identification number area ; and (j) the data extracted in step (d). Column (DS_P)
Is less than or equal to the data level value of the data string (DS_I) extracted in the step (c), the data extracted in the step (c) is stored in the data level value area of the data string (DS_P). Recording a value obtained by adding 1 to the data level value of the column (DS_I).

In the link information adding method according to the third aspect of the present invention, the procedure for extracting the data string in step (f) is the same as the search pattern 1 or the search pattern described in the data searching method according to the first aspect of the present invention. It can be similar to pattern 2. In steps (e) and (f), the data string extracted in step (d) is added to the parent identification number area constituting the parent data area of the data string (DS_I) extracted in step (c). It is determined whether or not there is link information related to (DS_P). This determines the element data area of the data string (DS_P) extracted in step (d).
This is equivalent to determining whether or not link information relating to the data string (DS_I) extracted in step (c) exists in the element identification number area to be formed .

In the link information adding method according to the third aspect of the present invention, in the step (f), the parent data area constituting the parent data area of the data string (DS_P) in which the input parent concept name is recorded. A data string is extracted based on the link information recorded in the identification number area . On the other hand, in the link information adding method according to a fourth aspect of the present invention described below, in step (f ′), the element data area of the data string (DS_I) in which the input data identification name is recorded is configured.
A data string is extracted based on the link information recorded in the element identification number area to be formed .

That is, in order to achieve the above second object,
The link information adding method according to the fourth aspect of the present invention is characterized in that
Steps of the link information adding method according to the third aspect of the present invention
(E ′) Instead of (e), (f), (g) and (h), (e ′) the data string (DS) extracted in the step (d)
_P) element data areaElement identification number area that constitutes
The data string (DS_) extracted in the step (c)
If link information relating to I) exists, the link information
(F ') The data string (DS) extracted in the step (d) is stopped.
_P) element data areaElement identification number area that constitutes
The data string (DS_) extracted in the step (c)
If there is no link information on I), step (c)
Element data area of data string (DS_I) extracted in
Element identification number area that constitutesLink information recorded in
Extract a data string based on the element data of the extracted data string.
Data areaElement identification number area that constitutesPhosphorus recorded in
Element data area of the data string extracted based on the
Element identification number area that constitutesUntil the data is empty
(G ′) any of the data string extracted in the step (f ′)
Is the data string (DS_P) extracted in step (d)
If it matches, the addition of the link information is stopped, and (h ′) any of the data strings extracted in the step (f ′)
The data string (DS_P) extracted in step (d)
If does not match,If the number of parent identification number areas is insufficient
If it has, create a new parent identification number area
so,Of the data string (DS_I) extracted in step (c)
Parent data areaParent identification number areaAnd the process
The data string (DS_P) extracted in (d)
Recording the link information.

In the link information adding method according to the fourth aspect of the present invention, the procedure for extracting the data string in step (f ′) is the same as the search pattern 3 or the data pattern described in the data searching method according to the second aspect of the present invention. This can be the same as the search pattern 4. In the steps (e ′) and (f ′), the data string (DS_P) extracted in the step (d) is extracted in the element identification number area constituting the element data area in the step (c). It is determined whether or not there is link information related to the data string (DS_I), which is determined in the parent identification number area constituting the parent data area of the data string (DS_I) extracted in step (c). , Is equivalent to determining whether there is link information relating to the data string (DS_P) extracted in step (d).

In the link information adding method according to the third and fourth aspects of the present invention, a data identifier and a parent concept name indicating a set including the data identifier are input. On the other hand, in the link information adding methods according to the fifth and sixth aspects of the present invention described below, a data identification name and an element concept name which is an element of a set represented by the data identification name are input.

According to a fifth aspect of the present invention, there is provided a link information adding method for achieving the second object, wherein the element identification number constituting the element data area of the data string (DS_I) is provided .
The present invention relates to a link information adding method for adding new link information relating to a data string (DS_E) corresponding to an element concept to an area . That is, in the link information adding method according to the fifth aspect of the present invention, each data string (DS_1) includes: (a) a data identification name area in which a data identification name is recorded; and (b) a data level value. (C) a parent identification number area for recording link information relating to a data string (DS_2) in which a data identifier indicating a set including the data identifier is recorded;
The parent data area, which has or more than the number of
And (d) an element identification number area for recording link information relating to a data string (DS_0) recording a data identification name which is an element of a set represented by the data identification name,
As many or more the number of (DS_0), main <br/> raw data area with, is composed of, from the data level value, the data level values of the data sequence (DS_1) of (A) data strings (DS_2) (B) configures the parent data area of the data string (DS_1)
The data string (DS_1) is stored in any of the parent identification number areas.
The link information that does not include the link information relating to itself and that is recorded in the parent identification number area constituting the parent data area of the data string (DS_1) is all different. (C) The parent of the data string (DS_1) Configure the data area
Starting from the link information recorded in an arbitrary parent identification number area, the parent identification number area constituting the parent data area of the data string
Until the area becomes empty, the parent that constitutes the parent data area of the data string
When the data sequence is sequentially traced based on the link information recorded in the identification number area , the link information on the data sequence (DS_1) indicates the parent data area of any traced data sequence.
Not included in the parent ID number regions constituting the data level values of (D) a data string (DS_0) is smaller than the data level values of the data sequence (DS_1), element data area of the (E) data strings (DS_1) Make up
The data string (DS_
1) An element identification number which does not include link information relating to itself and constitutes an element data area of the data string (DS_1)
All the link information recorded in the signal area is different, and (F) constitutes the element data area of the data string (DS_1) .
Starting from the link information recorded in the arbitrary element identification number area, the element identifiers constituting the element data area of the data string
Until another number area is empty, the element data area of the data row
When the data sequence is sequentially traced based on the link information recorded in the constituent element identification number area , the data sequence (DS_1)
Links infos are not included in the element identification number field that constitutes the element data area of any data sequence followed, in a data string assembly comprising a plurality of data strings, the element data area of a data string Element identification number to configure
A link information adding method for adding new link information to a data area, comprising: (a) inputting a data identification name and inputting an element concept name which is an element of a set represented by the data identification name; (B) if the input data identifier matches the input element concept name, stop adding link information; and (c) if the input data identifier does not match the input element concept name. Extracting a data string (DS_I) in which the input data identification name is recorded, and (d) extracting a data string (DS_E) in which the data identification name corresponding to the input element concept name is recorded. (E) Data string (DS_I) extracted in step (c)
If link information relating to the data string (DS_E) extracted in the step (d) exists in the element identification number area constituting the element data area of (c), the addition of the link information is stopped, and (f) step (c) (DS_I)
If there is no link information on the data string (DS_E) extracted in the step (d) in the element identification number area constituting the element data area of (a), the data string (DS_I) extracted in the step (c) based on the constituting parent data area extracting data sequence based on the link information recorded in the parent ID number area, link information recorded in the parent ID number area constituting the parent data area of the extracted data sequence, Until the parent identification number area constituting the parent data area of the extracted data string becomes empty,
(G) adding link information if any of the data strings extracted in step (f) matches the data string (DS_E) extracted in step (d); And (h) if none of the data strings extracted in step (f) match the data string (DS_E) extracted in step (d), the number of element identification number areas is Shortage
If it is, create a new element identification number area
Then, link information on the data string (DS_E) extracted in the step (d) is recorded in the element identification number area constituting the element data area of the data string (DS_I) extracted in the step (c), In addition, (i) the data string (DS_E) extracted in step (d)
In the parent identification number area constituting the parent data area in
If the number of parent identification number areas is insufficient, a new
Generating a parent identification number area and recording link information on the data string (DS_I) extracted in step (c); and (j) data string (DS_E) extracted in step (d). )
Is greater than or equal to the data level value of the data string (DS_I) extracted in step (c), the data level value recorded in the data string (DS_E) extracted in step (d) is 1 Is recorded in the data level value area of the data string (DS_I) extracted in step (c).

In the link information adding method according to the fifth aspect of the present invention, the procedure for extracting the data string in step (f) is the same as the search pattern 1 or the search pattern described in the data searching method according to the first aspect of the present invention. It can be similar to pattern 2. In steps (e) and (f), the data string extracted in step (d) is added to the element identification number area constituting the element data area of the data string (DS_I) extracted in step (c). It is determined whether or not the link information on (DS_E) exists.
The parent data area of the data string (DS_E) extracted in
This is equivalent to determining whether or not link information on the data string (DS_I) extracted in step (c) exists in the parent identification number area to be configured .

In the link information adding method according to the fifth aspect of the present invention, in the step (f), the parent data area constituting the parent data area of the data string (DS_I) in which the input data identification name is recorded. A data string is extracted based on the link information recorded in the identification number area . On the other hand, in the link information adding method according to a sixth aspect of the present invention described below, in step (f ′), the element data area of the data sequence (DS_E) in which the input element concept name is recorded is set.
A data string is extracted based on the link information recorded in the constituent element identification number area .

That is, the link information adding method according to the sixth aspect of the present invention for achieving the second object is a step (e) of the link information adding method according to the fifth aspect of the present invention. , (F), (g), and (h), (e ′) the data string (DS) extracted in the step (d).
_E), if link information on the data string (DS_I) extracted in the step (c) exists in the parent identification number area constituting the parent data area of (_E), stop adding the link information; (f ′) The data string (DS) extracted in the step (d)
_E), if there is no link information on the data string (DS_I) extracted in the step (c) in the parent identification number area constituting the parent data area of the data string ( make up the component data area of DS_E)
The data string is extracted based on the link information recorded in the element identification number area, and the element data area of the extracted data string is extracted.
Based on the link information recorded in the element identification number area constituting the
(G ′) extracting a data string until the element identification number area becomes empty; and (g ′) selecting one of the data strings extracted in the step (f ′) from the data extracted in the step (c). Column (DS_I)
If it matches, the addition of the link information is stopped, and (h ′) any of the data strings extracted in step (f ′) is replaced with the data string (DS_I) extracted in step (c).
If they do not match, the number of element identification number areas is insufficient
If it is, a new element identification number area is generated
Above, the data string (DS_I) extracted in step (c)
Recording the link information on the data string (DS_E) extracted in the step (d) in the element identification number area constituting the element data area.

The structure of the data string aggregate in the link information adding method according to the sixth aspect of the present invention is the same as the structure of the data string aggregate in the link information adding method according to the fifth aspect of the present invention. Can be the same.

In the link information adding method according to the sixth aspect of the present invention, the procedure of extracting the data string in step (f ′) is the same as the search pattern 3 or the data pattern described in the data searching method according to the second aspect of the present invention. This can be the same as the search pattern 4. In the steps (e ′) and (f ′), the data string (DS_E) extracted in the step (d) is extracted in the parent identification number area constituting the parent data area in the step (c). It is determined whether or not link information on the data string (DS_I) exists. This is because the element data area of the data string (DS_I) extracted in the step (c) is determined.
This is equivalent to determining whether or not link information on the data string (DS_E) extracted in step (d) exists in the constituent element identification number area . In addition, "Parent data area is configured
The parent identification number area to be
Area), which may be referred to as `` the element data area
The element identification number area is hereinafter simply referred to as the "element data area."
It may be called.

In the data search method or the link information adding method of the present invention, the parent data area (or element data area) of the data string does not include the link information relating to the data string itself, and the parent data area of the data string itself. The link information recorded in the data string (or the element data area) is different from the link information recorded in the parent data area (or the element data area) of the data string (DS_1). Until the data area (or element data area) becomes empty, when the data string is sequentially traced based on the link information recorded in the parent data area (or element data area) of the data string, the data string (DS_
The link information relating to 1) is not included in the parent data area (or element data area) of any data string. Therefore, it is possible to completely prevent the possibility of falling into an infinite loop in the search, and completely eliminate the cyclic definition described later.

In the data search method according to the present invention, the input data identification name is based on the link information recorded in the parent data area (or element data area) of the data string recorded in the data identification name area. A column is extracted, and the data sequence is extracted until the parent data region of the extracted data sequence becomes empty based on the link information recorded in the parent data area (or element data area) of the extracted data sequence. Therefore, not only a data string corresponding to the parent concept (or element concept) but also a data string corresponding to the indirect parent concept (or indirect element concept) can be extracted.
Further, unlike the related art, it is not necessary to record the element name redundantly in the element field.

In the link information adding method of the present invention,
In the above-mentioned step (f) or (f ′), a data string is extracted, and link information relating to the data string in which the parent concept name (or element concept name) in which the extracted data string is input is recorded. Find out if. by this,
When link information is added, it is possible to completely prevent the possibility of falling into an infinite loop in the search, and completely eliminate the cyclic definition described later.

[0045]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described based on embodiments with reference to the drawings. A specific example of the link information adding method will be described.

In the data structure shown in Table 1, for example, the element name “flammable” is recorded in the element fields of the record “pencil” and the record “tree”. Also,
“Tree” is recorded in the element field of the record “desk”. Therefore, when a record containing the element name “flammable” in the element field is searched, not only the record “pencil” and the record “tree” are extracted, but also the record “desk” containing “tree” in the element field is extracted. If possible, it is not necessary to record the element name “flammable” in the element field of the record “desk”.

In the element fields of the record “pencil” and the record “motor brush”, the element name “conductive”, the element name “soft” and the element name “black” are recorded. All of the names are contained in the record "Graphite". Therefore, the element name “graphite” is recorded in the element fields of the record “pencil” and the record “motor brush”, and any one of the element names recorded in the record “graphite” (for example, the element name “black”) ) In the element field, the record "graphite"
If you can extract not only record "pencil" and record "motor brush", but also record "pencil"
The element name “conductive”, the element name “soft” and the element name “black” can be deleted from the element field of the record “motor brush”. In other words, it is not necessary to record the element name “conductive”, the element name “soft” and the element name “black” in the element fields of the record “pencil” and the record “motor brush”.

Similarly, in the element field of the record "pencil", the element name "insulator" and the element name "flammable" are recorded, and these element names are included in the record "tree". . Therefore, when a record containing any element name (for example, element name “flammable”) recorded in the record “tree” in the element field is searched, the record “pencil” containing “tree” in the element field is also retrieved. If it can be extracted, the element name “flammable” can be deleted from the element field of the record “pencil”. Further, in the element fields of the record “desk” and the record “pencil”, the element names of the record “tree” such as the element name “insulator”, the element name “plant”, and the element name “specific gravity 1 or less” are recorded. Eliminates the need.

As described above, if there is no need to duplicately record the element name in the element field, data can be input very efficiently. In addition, it is possible to summarize concepts at various various levels, such as a lower concept, a middle concept, and a higher concept. Table 2 shows the data string simplified in this way.

[0050]

[Table 2]

First, the background of the data search method and the link information adding method of the present invention will be described with reference to Table 2. Table 2
Is easiest to interpret as follows: That is, the concept “desk” is a set including {tree, office supplies, furniture} as elements. The concept “pencil” is a set of {wood, graphite, office supplies, writing utensils, cylinders} as elements.
The concept "tree" is {insulator, flammable, plant, specific gravity 1 or less}
Is a set whose elements are. The concept "graphite" is
It is a set whose elements are soft and black. Concept "mica"
Is a set with {insulator} as an element. The concept “motor brush” is a set of {graphite} elements.

[0052] Originally, the concept is defined and enforced by several properties. Therefore, listing the properties that define the concept makes it possible to specify the concept. A set in mathematics can be determined by listing the elements included in the set. Thus, sets are convenient as models for concepts. Sets can include sets as elements. That is, the concept "pencil" includes the set of concepts "tree" and "graphite" as its elements. Therefore, a set type database for constructing a concept based on a set will be described below.

[Definition of Concept, Element Set, and Element Concept] First, some terms used are defined, and various rules are explained together. First, the concept A is defined and expressed as follows. "Concept name A" {element concept name X, element concept name Y, element concept name Z, ...} where "concept name A" is the name of concept "A" and element concept name "X" , “Y”, “Z”,... Are the names of the concepts that are the elements constituting the concept “A”. {Element concept name X, element concept name Y, element concept name Z,...} Are called an element set. On the other hand, some concepts do not have element concept names. Such a concept means that the element set is an empty set. The element concept name is composed of concepts having the same “concept name”. Such a concept is called an element concept.

For example, the concept “desk” can be expressed as follows. “Desk” ｛tree, office supplies, furniture｝ Here, “desk” is a concept name, and “tree”, “office supplies”, and “furniture” are element concept names. {Tree, office supplies, furniture} is a set of elements. Further, the element concept names "tree", "office supplies" and "furniture" are the concepts "tree" having the same concept names "tree", "office supplies" and "furniture".
It is composed of "office supplies" and "furniture", and the concepts "tree", "office supplies" and "furniture" are element concepts of the concept "desk". The concept "tree" can be expressed as follows. "Tree" {insulator, flammable, plant, specific gravity 1 or less}

On the other hand, the concepts “office supplies” and “furniture” are concepts that do not have element concept names. That is, the element set is an empty set. Therefore, it can be expressed as “office supplies” ｛｝ “furniture” ｛｝.

[Definition of parent concept and parent set]
, The concept A is called the parent concept of the concept a, and is expressed as follows. A∋ _e a In addition, a set consisting of all of the parent concept of the concept a is referred to as a parent concept set of concepts a.

For example, the concept “tree” is an element concept of the concept “desk”, and the concept “desk” is a parent concept of the concept “tree”. Further, the concept “tree” is an element concept of the concept “pencil”, and the concept “pencil” is a parent concept of the concept “tree”. Further, a set including the concept “desk” and the concept “pencil”, which are all parent concepts of the concept “tree”, is a parent concept set of the concept “tree”. The relationship between the concept "tree", the concept "desk" and the concept "pencil" can be expressed as follows. Concept "tree" ∈ _e concept "desk" Concept "tree" ∈ _e concept "pencil"

[Definition of Indirect Parent Concept and Indirect Element Concept] When concept A is an element concept of concept B (BＢ _e A) and concept B is an element concept of concept C (C∋ _e B) , Concept C is called the indirect parent concept of concept A, and is expressed as follows. On the other hand, the concept A is called an indirect element concept of the concept C. In the case of C∋ _i A or B∋ _e A and D 概念_i B, the concept D is
Is the indirect parent concept of the concept A, is expressed as D∋ _i A. On the other hand, concept A is called an indirect element concept of concept D. Furthermore,
Also in the case of B∋ _i A and C∋ _i B, concept C is an indirect parent concept of concept A and is expressed as C∋ _i A. The relationship between the concept "insulator" and the concept "desk" can be expressed as follows. Concept "desk" ∋ _i concept "insulator"

[Level of Concept] For example, the concept “insulator” has no element set. That is, the element set is an empty set. “Insulator” ｛｝ A concept whose element set is an empty set is called a first-level concept, and the name of such a concept is called a first-level concept name.
For example, the concepts “insulator”, “flammable”, “plant”, and “specific gravity 1 or less”, which are the element concepts that make up the concept “tree”, do not currently have the properties that define them, This is a first-level concept that has no association.

Further, a second level concept, a third level concept,...
.., The n-th level concept is expressed as follows. 2nd level concept: a concept in which the element concept is composed of the 1st level concept 3rd level concept: a concept in which the maximum level of the element concept is 2 and which has at least one second level concept as the element concept n Level concept: The maximum level of the element concept is (n-1), and at least one of the (n-
1) Concept with level concept

For example, the concept “tree”, the concept “graphite” and the concept “mica” are the second-level concept, the concept “desk” and the concept “pencil” are the third-level concept.

The second level concept, the third level concept,.
..., the name of the n-th level concept is changed to the second level concept name,
It is called a level concept name,..., N-th level concept name.

[Change in Concept Level] The concept level may change when an element concept is deleted or added.
That is, if all of the (n-1) th level concepts that are the element concepts are deleted from the element set of the nth level concept, the level of the concept becomes the maximum of the element concepts in the element set remaining after the deletion. It changes to the concept of the level which added 1 to the level. For example, if the element concept "tree" is deleted from the concept "desk", the concept "desk" changes from the third level concept to the second level concept. Conversely, when a concept at or above the nth level is added to the nth level concept as an element concept, the concept level changes to a concept of a level obtained by adding 1 to the maximum level of the element concept in the new element set. .

Even if the element set does not change, if the level of the element concept changes, the level of the parent concept or the indirect parent concept may change. For example, the concept “plant” is a first-level concept, but new concepts such as “life” and “composed of cells”
If the level concept becomes an element concept of the concept "plant", the concept "vegetation" changes to the second level concept. As a result, the concept "tree"
Although there is no change in the element concept that constitutes, the concept “tree” which is the parent concept of the concept “plant” changes from the second level concept to the third level concept. Along with this, the concepts "desk" and "pencil" which are indirect parent concepts of the concept "plant"
Changes from the third level concept to the fourth level concept.

[Prohibition of Duplication of Concept Name] Since the concept name is an identifier of the concept, the concept name must not be duplicated. Even if concept “X” and concept “Y” have the same element concept or element set, concept name “X” and concept name “Y” must be different.

[Construction Rules of Concept] Concept a and concept A are
When the following relationship is satisfied, the concept a can be made the element concept of the concept A. Rule: Concept a is not an element concept of concept A. Rule: Concept a is different from concept A. Rule: Concept a is neither an indirect parent concept of concept A nor a parent concept.

The rules are defined so that even if the same concept is added to the element set a plurality of times, the concept does not change and the operation becomes meaningless. Rules and rules are rules for eliminating the cyclic definition. If there is a rule, for example, a circular definition such as “desk” —desk, wood, office supplies, furniture, etc. can be eliminated. Further, when no rule exists, the concept "A" {B}, "B" {C}, and "C" {A}
, The concept “A” is indirectly circularly defined.

The following rules may be provided instead of the rules. Rule ': Concept A is not an indirect element concept of concept a,
And it is not an element concept.

[Basic Operations] The following four basic operations are allowed in a set type database for constructing a concept based on a set. (1) Generation of a new data string (2) Addition of link information for adding new link information to a predetermined area of the data string (3) Deletion of link information for deleting link information from a predetermined area of the data string (4) ) Data column disappearance

Next, the data search method and the link information adding method of the present invention will be described based on specific examples. Example 1
Example 3 to Example 14 relate to the data search method according to the first aspect of the present invention, and Example 4 and Example 15
The present invention relates to a link information adding method according to a first aspect of the present invention. Example 5 and Example 6 relate to the data search method according to the second aspect of the present invention, and Example 7 relates to the link information adding method according to the second aspect of the present invention. Example 8
The ninth and fourteenth embodiments relate to the data search method according to the third aspect of the present invention. Further, the tenth embodiment relates to a link information adding method according to the third aspect of the present invention, and relates to the first embodiment.
1 relates to a link information adding method according to a fourth aspect of the present invention, Example 12 relates to a link information adding method according to a fifth aspect of the present invention, and Example 13 relates to a sixth aspect of the present invention. The link information adding method according to the above.

Example 1 Example 1 relates to a data search method according to the first aspect of the present invention. In the first embodiment, a data string corresponding to the parent concept and / or a data string corresponding to the indirect parent concept is extracted from the data identifier to be searched.

The structure of the data string in the first embodiment is shown in FIG.
(A). The data string includes a data identifier name area (Name) in which a data identifier is recorded, a data level value area (Level) in which a data level value is recorded, and a parent data area (Parent). The data string further includes an identification number area (ID) in which an identification number for identifying the data string is recorded, and this identification number is a number unique to each data string.

The maximum data level value among the data level values recorded in the data string in the data structure is hereinafter simply referred to as the maximum data level value, and the maximum data level value is stored in the data level value area (Level). The recorded data sequence is called an Nth level data sequence. Also, data level value area (Level)
A data string in which a value of 1 is recorded in the data level is referred to as a first level data string.
A data string in which the value of <L <N) is recorded is referred to as an Lth level data string. Further, the data identifiers recorded in the data identifier areas of the Nth level data sequence, the Lth level data sequence, and the first level data sequence are respectively referred to as an Nth level data identifier, an Lth level data identifier, It is called the first level data identifier. The data identification name corresponds to the concept name described above.

In the data structure according to the first embodiment, the data level value area (Level) of each data string includes:
It is assumed that the value of 1, 2, or 3 is recorded. That is,
The data structure according to the first embodiment includes a plurality of first-level data strings, a plurality of second-level data strings, and a plurality of third-level data strings.

The Nth level data identifier is a name indicating a set of at least one (N-1) th level data identifier. Alternatively, the Nth level data identifier comprises at least one (N-1) th level data identifier;
In addition, one or more L'th level data identifiers (where L '
= 1, 2,..., N−1). The L-th level data identifier is a name indicating a set including one (L-1) th level data identifier. Alternatively, the L-th level data identifier is a name indicating a set including at least one (L-1) th level data identifier. Alternatively, the L-th level data identifier comprises at least one (L-1) th level data identifier, and further includes one or more L-level data identifiers (where L "=
1, 2,..., L-1). These data identification names are recorded in the data identification name area (Name) of each data string.

The parent data area (Parent) of each data string (DS_1) relates to a data string (DS_2) recording a data identification name indicating a set including the data identification name recorded in the data string (DS_1). Link information is recorded. The parent data area (Parent) is composed of a plurality of parent identification number areas (Parent_ID). The data level value of the data string (DS_2) corresponding to the link information recorded in the parent data area (Parent) is larger than the data level value of the data string (DS_1) including the parent data area (Parent).

The M-th level data identifier (where M = 1,
, M−1 level data identifiers (M ′ = M + 1, M + 2,.
N), the M'th level data string (DS_2)
Link information is recorded in the parent data area (Parent) of the M-th level data string (DS_1). Specifically, the identification number of the M'th level data string (DS_2) is recorded in the parent identification number area (Parent_ID) of the parent data area (Parent) in the Mth level data string (DS_1). The identification number of the M'th level data string (DS_2) recorded in the parent identification number area (Parent_ID) of the parent data area (Parent) in the Mth level data string (DS_1) is as follows:
9 is link information according to the first embodiment.

With reference to the M-th level data string (DS_1), the M′-th level data string (DS_2) corresponds to the parent concept. Further, the data identifier recorded in the M'th level data sequence (DS_2) corresponds to the parent concept name described above. On the other hand, based on the M'th level data string (DS_2), the Mth level data string (DS_1) corresponds to the element concept. When the data identifier recorded in the M'th level data string (DS_2) is used as a reference, the data identifier recorded in the Mth level data string (DS_1) corresponds to the element concept name. Furthermore, the parent identification number area (Parent_I) of the parent data area (Parent) of the M-th level data sequence (DS_1)
The data identifier of the M'th level data string (DS_2) corresponding to the link information recorded in D) corresponds to a set of parent concepts.

Table 3 shows a specific example of a data identifier corresponding to a concept name and its data level value, and a data identifier corresponding to an element concept name and its data level value in the data structure of the first embodiment. Show.

[0080]

[Table 3]

FIG. 5B shows a specific data string structure of the first level data string indicated by ID number 5 in Table 3. The data identification name in this first level data string is “insulator”, and this data identification name is recorded in the data identification name area (Name) of the data string. In the data level value area (Level_Value), “1” indicating that the data string is the first level data string is recorded.

Further, since there is an M′-th level data identifier (M ′ = 2 or 3 in this case) indicating a set including the first-level data identifier “insulator” (specifically, The two-level data identifiers “tree” and “mica”), and the identification numbers 12 and 14 of the M′-th level data sequence in which these M′-level data identifiers are recorded are the first level identified by ID number 5 It is recorded as link information in the parent identification number area (Parent_ID) of the parent data area (Parent) in the data string. ID numbers 1-4, 6-11, which are the first level data strings
Has a similar structure. In some cases, there is an Mth level data string in which there is no M'th level data identifier indicating a set including the Mth level data identifier. In this case, the parent data area (Pa
Nothing is recorded in the parent identification number area (Parent_ID) of the rent).

In Table 3, the second (=) indicated by ID number 12
M) A specific data string structure of the level data string is illustrated in FIG. The second in this second level data sequence
The data identifier that is the level data identifier is “tree”,
This data identifier is the data identifier name area (Na
me). The data level value area (Leve
In “l_Value”, “2” indicating that the data string is the second level data string is recorded.

The M'th level data identifier which is the name of the set including the data identifier "tree" (M '= 3 in this case)
(Specifically, the data identifiers “desk” and “pencil”), the identification numbers 15 and 16 of the M′-th level data sequence in which these M′-level data identifiers are recorded are IDs
It is recorded as link information in the parent identification number area (Parent_ID) of the parent data area (Parent) in the second level data string indicated by number 12.

ID number 13, 1 which is the second level data string
4 has a similar structure.

In Table 3, the third (=) indicated by ID number 15
N) A specific data string structure of the level data string is illustrated in FIG. The data identification name in this third level data string is “desk”, and this data identification name is recorded in the data identification name area (Name) of the data string. In the data level value area (Level_Value), “3” indicating that the data string is the third level data string is recorded. In the first embodiment, since the maximum data level value (N) of the data string is 3, the parent identification number area (Parent_I) of the parent data area (Parent) in the third level data string is used.
Nothing is recorded in D). The ID numbers 16 and 17, which are the third level data string, have the same structure.

In Table 3, ID No. 16 and ID No. 1
The specific data string structures of the third level data string and the second level data string indicated by 4 are illustrated in FIGS. 7A and 7B, respectively.

To eliminate the cyclic definition, the data string (DS
_1) does not include link information (for example, an identification number) relating to the data string (DS_1) itself. Moreover, the data string (DS_1)
Are different from each other in the parent data area (Parent). These are the rules described above [Construction rules of concept]
Rules and rules.

Further, starting from any link information recorded in the parent data area of the data string (DS_1), the link recorded in the parent data area of the data string until the parent data area of the data string becomes empty. When the data sequence is sequentially traced based on the information, the link information on the data sequence (DS_1) is not included in the parent data area of any traced data sequence. This corresponds to the rule of [Construction rules of concept] described above.

For example, the ID number 12 which is the link information recorded in the parent identification number area (Parent_ID) of the parent data area (Parent) of the first level data string (DS_1) indicated by the ID number 5 is changed to an arbitrary link. When the information is information, the data sequence is sequentially traced from the link information based on the link information recorded in the parent data area of the data sequence. That is, I
Based on the ID number 15 which is the link information recorded in the parent identification number area (Parent_ID) of the parent data area (Parent) of the data string of D number 12, the data string of ID number 15 is reached. Parent data area (Pare
The parent identification number area (Parent_ID) of (nt) is empty. Therefore, next, the parent data area (Pa
rent), based on the ID number 16 which is the link information recorded in the parent identification number area (Parent_ID) next to the ID number 1
6 is reached. Parent identification number area (Parent_ID) of parent data area (Parent) of data string of ID number 16
Is empty. Therefore, next, the parent identification number area (Parent) next to the parent data area (Parent) of the data string of ID number 12
_ID), the parent identification number area (Parent_ID) is empty. In such processing, the ID number 5 which is the link information on the data string (DS_1) corresponds to any data string (in this example, ID
It is not included in the parent data area of the data strings of numbers 12, 15, and 16).

When the first level data string indicated by ID number 5 is used as a reference, the second level data string of ID number 12 corresponds to the parent concept, and the third level data string of ID number 15 is indirect. It corresponds to the parent concept. Conversely, when the third level data string indicated by ID number 15 is used as a reference, the second level data string of ID number 12 corresponds to the element concept, and the first level data string of ID number 5 is indirect. It corresponds to the element concept.

FIG. 34 is a conceptual diagram of an apparatus suitable for executing the data search method or the link information adding method of the present invention. This device includes a processing device, an input device 12, a display device 1
6, an external storage device 18 such as a main storage device 14 and a hard disk. The processing device includes, for example, a microprocessor unit 10 (hereinafter, abbreviated as MPU 10). The input device 12 includes, for example, a pointing device (mouse) and a keyboard. The display device 16 is a cathode ray tube (CRT)
, And various data are displayed. Main storage device 1
4 includes, for example, a random access memory (RAM), and stores programs to be processed by the MPU 10, various data, data strings, and the like. The external storage device 18 stores a data structure to be subjected to the data search method or the link information adding method of the present invention. The data structure is
It can be in any file format.

In the data search method according to the first embodiment, first, a data identifier corresponding to data to be searched is input. Next, a data string is extracted based on the link information recorded in the parent data area of the data string in which the input data identification name is recorded in the data identification name area (hereinafter, may be referred to as an input data string). And further, based on the link information recorded in the parent data area of the extracted data string,
The data string is extracted until the parent data area of the extracted data string becomes empty, and the extracted data string is output.
The search pattern of the data search method in the first embodiment is set to the search pattern 1 shown in FIG.

The data search method according to the first embodiment will be described below with reference to FIG.
This will be described with reference to the flowchart of FIG.

As shown in FIG. 8, first, a character string "insulator" is inputted as a data identification name corresponding to the data to be searched from the input device 12 composed of, for example, a keyboard (step S10). The MPU 10 reads the input character string “insulator” and stores it in a variable area (not shown) in the main storage device 14. Next, the MPU 10 accesses the data structure stored in the external storage device 18 and stores the data structure in which the data identification name recorded in the data identification name area (Name) of the data sequence matches the input character string. A column (which is an input data sequence, and in the first embodiment, a data sequence of ID number 5) is read from the data structure (step-S11), and the contents of the input data sequence are stored in a variable area in the main storage device 14. Remember. If the reading of the data string from the data structure has failed, the search processing ends (step-S12).

Next, the MPU 10 checks the parent identification number area of the parent data area of the input data string stored in the variable area in the main storage device 14. If the parent identification number area is empty, the MPU 10 terminates the search processing. (Step-S13). If the parent identification number area is not empty, the MPU 10 prepares a dynamic array D (0) in the variable area and initializes the dynamic array D (0) (Step-S14). Note that the dynamic array D stores the ID numbers of the data strings. Then, the ID of the input data string
The number is stored in the dynamic array D (0) (step-S1
5).

Next, a data string is extracted based on the link information recorded in the parent data area of the input data string, and the extracted data string is extracted based on the link information recorded in the parent data area of the extracted data string. Data strings are extracted until the parent data area becomes empty.

Specifically, search processing 1 shown in FIG. 9 is executed (step-S16). In the search process 1, first, the variable i is set to 0 (Step-S20).

Then, the MPU 10 has the dynamic sequence D (i)
Is read from the data structure and stored in the variable X (step-S2).
1). Note that the structure of the variable X can be the same as the structure of the data string. Next, the parent identification number area (Parent_ID_X) of the parent data area (Parent_X) in the variable X is checked. If the parent identification number area (Parent_ID_X) is empty, the process proceeds to step S28 (step S22). If the parent identification number area (Parent_ID_X) is not empty, the ID number stored in the first parent identification number area (Parent_ID_X) is stored in a variable PID (step-S23).

Next, the MPU 10 checks whether or not the ID number stored in the variable PID exists in the dynamic array D (step-S24).
Move to 6. On the other hand, if it does not exist, the variable PID
Is added to the end of the dynamic array D (step-S25). Next, the MPU 10 checks a parent identification number area (Parent_ID_X) next to the parent data area (Parent_X) in the variable X, and checks the parent identification number area (Parent_ID_X).
If (ID_X) is empty, the process moves to step S28 (step S26). If the parent identification number area (Parent_ID_X) is not empty, the ID number stored in the parent identification number area (Parent_ID_X) is stored in a variable PID (step-
S27), and return to step S24.

In step S28, the value of the variable i is incremented by one (step S28), and then whether the ID number is stored in the dynamic array D (i)
That is, it is determined whether or not the dynamic array D (i) is at the end.
Return to 21. If the ID number is not stored in the dynamic array D (i), the search processing 1 ends (step-S2).
9).

Thereafter, the MPU 10 executes the dynamic array D (i)
The data string corresponding to the ID number stored in the data structure is read from the data structure, and the data identification name recorded in the data identification name area of the data string and the data level value recorded in the data level value area are displayed, for example, on the display device 16. (Step-S17). The data identifier to be searched may or may not be displayed. Thus, the search processing ends.

When "insulator" is searched, the ID numbers are stored in the dynamic array D (i) in the order shown in Table 4 below.

[0104]

[Table 4] D (0): ID number = 5 D (1): ID number = 12 D (2): ID number = 14 D (3): ID number = 15 D (4): ID number = 16

Thus, in the first embodiment, when “insulator” is input as the data identifier to be searched, the second level data string in which the second level data identifier indicating the set including “insulator” is recorded Not only are “trees” and “mica” extracted, but also “desks” and “pencils”, which are third-level data strings that record third-level data identifiers indicating sets that include “trees”. Can be extracted. That is, "insulator" is added to "desk" and "pencil" which are the third level data strings.
It is possible to derive that the concept of "insulator" is included in "desk" and "pencil" without inputting the element "."

Here, based on the first-level concept “insulator”, the second-level concepts “tree” and “mica” correspond to the parent concept, and the third-level concept “desktop”. ] And “pencil” correspond to the indirect parent concept. The search results are shown in Table 5 below. For reference, after the data level value,
It indicates whether it is a parent concept or an indirect parent concept.

[0107]

[Table 5] Data level value of extracted data sequence Concept classification Data identification name Tree 2 Parent concept Mica 2 Parent concept Desk 3 Indirect parent concept Pencil 3 Indirect parent concept

In the first embodiment, the first level data identifier is searched, but the second to (N-1) th level data identifiers are searched depending on the value of N. Can also be targeted. Further, the output of the extracted data string may be performed at the time when the ID number stored in the variable PID is added to the end of the dynamic array D in step-25.

(Embodiment 2) Embodiment 2 is a modification of Embodiment 1. In the first embodiment, the search pattern of the data search method is the search pattern 1 shown in FIG.
In, the search pattern of the data search method was set to search pattern 2 shown in FIG. The data search method according to the second embodiment will be described below with reference to the flowcharts in FIGS.

As shown in FIG. 10, first, a character string “insulator” is input as a data identification name corresponding to data to be searched from the input device 12 composed of, for example, a keyboard (step S30). The MPU 10 reads the input character string “insulator” and stores it in a variable area set in the main storage device 14. Next, the MPU 10 accesses the data structure stored in the external storage device 18,
A data string in which the data identification name recorded in the data identification name area (Name) of the data string matches the input character string (an input data string, and in the second embodiment, a data string of ID number 5) read from the data structure (step -S31), the content of such input data sequence, is stored in the variable Y ₁ of the main storage device 14. The structure of the variables Y ₁ , Y ₂ ,... Can be the same as the structure of the data string.
If the reading of the data string from the data structure has failed, the search processing ends (step-S32).

Next, the MPU 10 controls the parent data area (Par) of the input data string stored in the variable Y ₁ in the main storage device 14.
The parent identification number area (Parent_ID_Y) of the parent identification number area (Parent_ID_Y) is read out (step-S33).
If (Y) is empty, the search processing is terminated (step-S3).
4). In the second embodiment, as shown in FIG. 5B, the parent data area (Pare
Since the ID number 12 is recorded as link information in the parent identification number area (Parent_ID) at the head of (nt), search processing 2 is executed (step-S35).

In search process 2, the routine shown in FIG. 11 is mainly executed. In the following description of the second embodiment, in order to help understand the execution procedure of the routine, FIG.
See FIG. In FIG. 12, the subscript after “R” is
The order of execution of the routine is shown, and the numbers in the circles indicate the ID numbers of the data strings. In the routine, based on the ID number 12, the MPU 10 accesses the data structure stored in the external storage device 18, reads the data string of the ID number 12 from the data structure, and reads the variable Y in the main storage device 14. ₂ stores the contents of the data string of ID number 12. Then, "tree" which is the second level data identifier recorded in the data identifier area (Name) of the data string of ID number 12, and "2" which is the number of levels recorded in the data level value area (Level_Value) Is displayed on the display device 16 (R ₁ ).

Next, the routine is repeated. That is, MPU
10 is the ID stored in the variable Y ₂ in the main storage device 14.
The parent data area (Pare
The data stored in the parent identification number area (Parent_ID_Y) at the head of (nt_Y) is read as link information. In the second embodiment, as shown in FIG. 6A, the ID number 15 is recorded in the parent identification number area (Parent_ID) at the head of the parent data area (Parent) in the data string of ID number 12. I have. Based on the ID number 15, the MPU 10 accesses the data structure stored in the external storage device 18, reads the data string with the ID number 15 from the data structure, and stores the ID number in the variable Y ₃ in the main storage device 14. The contents of the 15 data strings are saved. Then, "desk" which is the third level data identification name recorded in the data identification name area (Name) of the data string of ID number 15, and the data level value area (Leve)
The level number “3” recorded in (l_Value) is displayed on the display device 16 (R ₂ ).

Further, the MPU 10 stores the data in the content of the data string with the ID number 15 stored in the variable Y ₃ in the main storage device 14 in the first parent identification number area (Parent_ID_Y) of the parent data area (Parent_Y). Read data as link information. In the second embodiment, as shown in FIG. 6B, nothing is recorded in the parent identification number area (Parent_ID) at the head of the parent data area (Parent) in the data string of ID number 15. Therefore, the process exits from the routine (R ₃ ) and discards the variable Y ₃ in the main storage device 14.

Further, the search processing 2 is repeated. That is, MP
U10 is the I stored in the variable Y ₂ in the main storage device 14.
The parent data area (Pa) in the content of the data string of D number 12
The data stored in the parent identification number area (Parent_ID_Y) next to (rent_Y) is read as link information. In the second embodiment, as shown in FIG. 6A, the ID number 16 is recorded in the second parent identification number area (Parent_ID) of the parent data area (Parent) in the data string of ID number 12. ing. Therefore, based on ID number 16, MPU1
0 accesses the data structure stored in the external storage device 18 reads the data sequence ID number 16 from the data structure, the main memory device 14 variables Y ₃ in ID number 1
Save the contents of the data string of No. 6. And ID number 16
"Pencil" which is the third level data identifier recorded in the data identifier area (Name) of the data string, and "3" which is the number of levels recorded in the data level value area (Level_Value)
Is displayed on the display device 16 (R ₄ ).

[0116] Then, MPU 10 is stored in the head of the parent identification number field of the parent data area in the content of the main memory 14 the data sequence ID number 16, which is stored in the variable Y ₃ in (Parent_Y) (Parent_ID_Y) The read data is read as link information. In the second embodiment, as shown in FIG. 7A, nothing is recorded in the parent identification number area (Parent_ID) at the head of the parent data area (Parent) in the data string of ID number 16. Therefore, the routine exits (R ₅ ) and discards the variable Y ₃ in the main memory 14.

Next, the MPU 10 is stored in the parent identification number area (Parent_ID_Y) next to the parent data area (Parent_Y) in the content of the data string of ID number 12 stored in the variable Y ₂ in the main storage device 14. Read the data. In the second embodiment, as shown in FIG. 6A, nothing is recorded in the third parent identification number area (Parent_ID) of the parent data area (Parent) in the data string of ID number 12. . Therefore, it exits from the routine (R ₆ ) and discards the variable Y ₂ in main memory 14.

[0118] Once the search is terminated 2 (step -S35), then as shown in FIG. 10, MPU 10 is a main storage device 14 in the variable Y ₁ has been data sequence ID number 5 stored in The data stored in the parent identification number area (Parent_ID_Y) next to the parent data area (Parent_Y) in the content is read as link information (step-S36), and the process returns to step-S34.

In the second embodiment, as shown in FIG. 5B, the ID number 14 is assigned to the second parent identification number area (Parent_ID) of the parent data area (Parent) in the data string of ID number 5. Is recorded as link information.
Therefore, the data string is extracted again based on the link information recorded in the parent data area of the input data string, and the parent of the extracted data string is extracted based on the link information recorded in the parent data area of the extracted data string. Until the data area is empty
Extract a data string.

That is, search processing 2 is executed again (step-S35). In other words, the MPU 10
Based on the following conserved ID number in the parent ID number area (Parent_ID_Y) of the parent data area (Parent_Y) in the context of data sequence ID No. 5, which is stored in the variable Y ₁ of the inner, M
PU10 accesses the data structure stored in the external storage device 18 reads the data sequence ID number 14 from the data structure, ID to a variable Y ₂ of the main storage device 14
The contents of the data string of No. 14 are saved. Then, “mica” which is the second level data identifier recorded in the data identifier area (Name) of the data string of ID number 14 and “2” which is the number of levels recorded in the data level value area (Level_Value) Is displayed on the display device 16 (R ₇ ).

Next, the MPU 10 is stored in the parent identification number area (Parent_ID_Y) at the head of the parent data area (Parent_Y) in the content of the data string of ID number 14 stored in the variable Y ₂ in the main storage device 14. Read the data. In the second embodiment, as shown in FIG. 7B, nothing is recorded in the parent identification number area (Parent_ID) at the head of the parent data area (Parent) in the data string of ID number 14. Therefore, the process exits from the routine (R ₈ ) and discards the variable Y ₂ in the main storage device 14.

[0122] Once the next search processing 2 (Step -S35) Exit, MPU 10 is the parent data area in the content of the data string of the ID number 5 is stored in the variable Y ₁ of the main storage device 14 (Parent_Y) Parent identification number area (Parent_ID_
Read the data stored in (Y) (Step-S3)
6), and return to step S34. However, in the second embodiment, as shown in FIG. 5B, nothing is recorded in the third parent identification number area (Parent_ID) of the parent data area (Parent) in the data string of ID number 5. Not. Therefore, all search processing ends.

Table 6 below shows the display of the search results in the second embodiment. For reference, after the data level value, the parent concept or the indirect parent concept is shown.

[0124]

[Table 6] Data level value of extracted data sequence Concept classification Data identification name Tree 2 Parent concept Desk 3 Indirect parent concept Pencil 3 Indirect parent concept Mica 2 Parent concept

Embodiment 3 Embodiment 3 also relates to a data search method according to the first aspect of the present invention. In the first and second embodiments, one data identifier to be searched is used. On the other hand, in the third embodiment, there are a plurality of data identifiers to be searched. Then, the data search method described in the first or second embodiment is executed for each data identifier to be searched. Then, the number of data identifiers to be searched that are related to each of the extracted data strings is output. Hereinafter, the data search method for outputting the number of data identifiers to be searched for each of the extracted data strings will be referred to as a scrap-type data search method.

More specifically, the third embodiment will be described using the data structure described in the first embodiment as an example. Although the search pattern is the same as the search pattern 1 in the first embodiment, the search pattern 2 described in the second embodiment can also be used. There were three data identifiers to search for, namely "insulator", "black" and "office supplies". Note that the data level values of the data strings in which the data identifiers to be searched are recorded may be the same or different. The data search method according to the third embodiment will be described below with reference to the flowcharts in FIGS.

As shown in FIG. 13, first, a character string "insulator" or "black" is input as a data identification name corresponding to the data to be searched from the input device 12 such as a keyboard.
And "Office Supplies" (step S40). M
The PU 10 reads the input character strings “insulator”, “black” and “office supplies” and stores them in a variable area (not shown) in the main storage device 14. Next, after substituting “1” for the variable j, the MPU 10 accesses the data structure stored in the external storage device 18 and records the data identification name (Name) recorded in the data identification name area (Name) of the data string. Example 3
, The data string (the first input data string) in which the input character string matches the input character string
Is read from the data structure, and the contents of the data string are stored in a variable area in the main storage device 14. This process is executed by the number of input data identifiers. That is,
Next, in the third embodiment, the MPU 10 accesses the data structure stored in the external storage device 18 and
The data string with the number 11 is read from the data structure, the contents of the data string (the second input data string) are stored in the variable area in the main storage device 14, and the MPU 10
Accesses the data structure stored in the external storage device 18, reads the data string of ID number 1 from the data structure, and stores the contents of the data string (third input data string) in the main storage device. 14 is stored in a variable area (step-S41). If the reading of the data string from the data structure has failed, the search process ends (step-S42). Next, the MPU 10 checks the parent identification number area of the parent data area of all the data strings stored in the variable area in the main storage device 14. If the parent identification number areas of all the data strings are empty, the MPU 10 performs the search processing. (Step-S4)
3).

On the other hand, if the parent identification number area of any of the input data strings is not empty, MPU 10 prepares dynamic array D (0) in the variable area and initializes dynamic array D (0). (Step-S44). Note that the dynamic array D stores the ID numbers of the data strings. After that, among the data strings stored in the variable area in the main storage device 14, the ID number 5 of the first input data string which is the data string in which the first data identifier to be searched is recorded is assigned to the dynamic array D. (0) (Step-S45).

Next, a data string is extracted based on the link information recorded in the parent data area of the first input data string, and further based on the link information recorded in the parent data area of the extracted data string. Data strings are extracted until the parent data area of the extracted data string becomes empty. Specifically, a search process 1 similar to the first embodiment shown in FIG. 9 is executed (step-S46).

Thereafter, the MPU 10 reads the sequence D _j (= D ₁ )
Is prepared, the contents of the dynamic array D (i) are copied to the array D _j (i), and the dynamic array D (i) is discarded (step-S4).
7). Next, the MPU 10 checks whether or not the next input data string is stored in the variable area in the main storage device 14 (Step-S48). If the next input data string is not stored in the variable area in the main storage device 14, step S5
Move to 0. Incidentally, the content of the D _1, shown in Table 7 below.

[0131]

[Table 7] D ₁ (0): ID number = 5 D ₁ (1): ID number = 12 D ₁ (2): ID number = 14 D ₁ (3): ID number = 15 D ₁ (4): ID number = 16

If the next input data string is stored in the variable area in the main storage device 14, the variable j is incremented by one, and the process returns to step S44. That is, MP
U10 prepares the dynamic array D (0) in the variable area and initializes the dynamic array D (0) (Step-S44). Thereafter, the ID number 11 of the second input data string is stored in the dynamic array D (0) (Step-S45).

Next, a data string is extracted based on the link information recorded in the parent data area of the second input data string, and further based on the link information recorded in the parent data area of the extracted data string. Data strings are extracted until the parent data area of the extracted data string becomes empty. Specifically, search processing 1 similar to that of the first embodiment is executed (step-S46).

Thereafter, the MPU 10 reads the sequence D _j (= D ₂ )
Is prepared, and the contents of the dynamic array D (i) are copied to the array D ₂ ,
The dynamic array D (i) is discarded (Step-S47).
Incidentally, the content of D _2, shown in Table 8 below.

[0135]

[Table 8] D ₂ (0): ID number = 11 D ₂ (1): ID number = 13 D ₂ (2): ID number = 16 D ₂ (3): ID number = 17

Next, the MPU 10 checks whether the next input data string is stored in the variable area in the main storage device 14 (step S48). In the third embodiment, the next input data string (ID number 1) is stored in the variable area in the main storage device 14.
After the variable j is incremented by one, the process returns to step S44 to execute steps S45, S46, and S47. In step S47, the contents of the dynamic array D (i) are copied to the array D _j (= D ₃ ). Note that D
The contents of ₃ are shown in Table 9 below.

[0137]

[Table 9] D ₃ (0): ID number = 1 D ₃ (1): ID number = 15 D ₃ (2): ID number = 16

Next, the MPU 10 checks whether the next input data string is stored in the variable area in the main storage device 14 (step S48). In the third embodiment, since the next input data string is not stored in the variable area in the main storage device 14, the process proceeds to step S50.

In step S50 shown in FIG.
The MPU 10 prepares and initializes the dynamic array R (0).
This dynamic array R includes an ID number area (ID_R) and a count area (Count_R).

Next, the MPU 10 sets “1” in the variables i and j.
Is substituted (step-S51), and the contents (ID number) of the array D _j (i) are read (step-S52). Then, MPU 10 checks whether the data is not present in the sequence D _j (step -S53), in the absence if proceeds to step -S58. If it exists, it is checked whether or not the contents (ID numbers) of the array D _j (i) are recorded in the entire ID number area (ID_R) of the dynamic array R (step-S54). If the ID number is not recorded, the ID number area (ID_R) recording the ID number and the count area (Count_R) recording “1” are recorded.
R) is added to the end of the dynamic array R (step-S5).
5). If it has been recorded, the count area (Count_Count) of the element of the dynamic array R in which the ID number is recorded.
R) is incremented by one (step S5).
6). Next, the variable i is incremented by one (step S57), and the process returns to step S52.

In step S58, the MPU 10
Substitutes "1" for the variable i, increments the variable j by 1, and checks whether the value of the variable j has exceeded the number of input data strings (step-S59). If the value of the variable j does not exceed the number of input data strings, the process returns to step S52. If the value of the variable j exceeds the number of input data strings, MP
U10 reads a data string corresponding to the ID number recorded in the ID number area (ID_R) from the data structure in the main storage device 14 in order from the dynamic array R (1), and reads the data identifier of the data string. The data identifier recorded in the area,
And the value (the number of hit input data strings) recorded in the count area (Count_R) is displayed (step-S6).
0). A display example is shown in Table 10 below. For reference, the input data identifier to be searched is attached after the number of hit input data strings.

[0142]

[Table 10] Extracted data Hit input Data identifier of the column to be searched Number of data columns Data identifier Tree 1 Insulator Mica 1 Insulator Desk 2 Insulator / Office Supplies Pencil 3 Insulator / Black / Office Supplies Graphite 1 Black Motor brush 1 Black

In the third embodiment, the number of data identifiers to be searched (the number of hit input data strings) related to each of the extracted second-level data string and third-level data string is output. You. It should be noted that what data level value of the extracted data sequence has a data identifier to be searched for which is to be output is determined in advance by inputting the data level value from the input device 12. It should be left.

In such a data retrieval method of the scrap system, the count area (Count_R) of the dynamic array R
(Number of input data strings hit) indicates how many of the three data identifiers, “insulator”, “black” and “office supplies”, which are the data identifiers to be searched, are extracted data. Indicates whether the column contains or is related. For example, since the value of the number field in the third-level data string having “pencil” as the data identifier is “3”, the three data identifiers to be searched are directly added to the third-level concept name “pencil”. Included indirectly or indirectly.

Here, being directly included means that the concept represented by the data identifier to be searched is an element concept of a certain concept. On the other hand, indirectly included
It means that the concept represented by the data identifier to be searched is an indirect element concept of a certain concept. Specifically, the first
The level concept name "Office Supplies" is directly included in the third level concept name "Pencil". On the other hand, the first-level concept name “insulator” and “black” are indirectly included in the third-level concept name “pencil”.

Here, the third-level concept name “pencil” including all three first-level concept names to be searched is obtained by replacing the three first-level concept names to be searched with the conventional search condition “AND”. "
It is equivalent to the search result obtained when tied with.

For example, the count area (Coun- ter) in the third level data string having "desk" as a data identifier.
t_R) (the number of input data strings hit)
Is "2", the first level concept name corresponding to the two data identifiers to be searched is directly or indirectly included in the third level concept name "desk". The value (number of cases) recorded in the count area (Count_R) in the second level data string having "tree", "graphite" and "mica" as data identifiers is "1", and "motor brush" is Since the value (number of records) recorded in the count area (Count_R) in the third level data string having the identifier is “1”, the first level concept name corresponding to one data identifier to be searched is: These second-level concept names and third-level concept names are directly or indirectly included.

That is, the second or third-level concept name including at least one concept name corresponding to the three data identifiers to be searched is obtained by replacing the three concept names to be searched with the conventional search condition " This corresponds to a search result obtained when tied with “OR”. However, unlike the conventional data search method, it is possible to check how many data identifiers to be searched are related to each of the extracted second-level data string and third-level data string.

In the third embodiment, the data identifier recorded in the first level data string is searched. However, depending on the value of the level N, the search target is the first level data string to the first level data string. (N-1) The data identifiers recorded in the level data string may be mixed.

Example 4 Example 4 relates to a link information adding method according to the first aspect of the present invention. The structure of the data string in the data structure of the fourth embodiment and various data recorded in each data string were the same as in the first embodiment.
Hereinafter, a link information adding method according to the fourth embodiment will be described. Prior to the description, a process or an operation related to generation of a data string which is introduction of a new data string into a data structure will be described.

Preferably, the generation of a new data string in the data structure is performed by introducing a new first-level data string into the data structure. In this case, when a new data identifier is input from the input device 12, MP
U10 checks whether the input data identification name has already been recorded in the data identification name area (Name) of each data string in the data structure. If the input data identifier is one of the data identifier names (Nam
If it is already recorded in e), the MPU 10 stops the processing. Otherwise, the MPU 10 stores the data string having the structure shown in FIG.
4, a new ID number is assigned, the ID number is recorded in the identification number area (ID), the input data identification name is recorded in the data identification name area (Name), and "1" is set. Record in the data level value area (Level). Then, this data string is added to the data structure in the external storage device 18.

The link information adding method according to the fourth embodiment will be described below with reference to the flowcharts in FIGS.

As shown in FIG. 15, first, a data identification name is input from the input device 12 composed of, for example, a keyboard, and a parent concept name indicating a set including the data identification name is input (step-). S70). In Example 4, the data identification name was “flammable” and the parent concept name was “plant”. The MPU 10 reads the input character strings “flammable” and “plant” and stores them in a variable area set in the main storage device 14.

The MPU 10 checks whether the input data identification name matches the input parent concept name (step-S71). If the data identification name matches, the MPU 10 violates the above-described rule. To stop.

If the input data identification name does not match the input parent concept name, the data string (DS_I, input data string) in which the input data identification name is recorded is read out. Read a data string (DS_P, hereinafter sometimes referred to as a parent concept data string) in which a data identification name that matches the parent concept name is recorded (step-
S72). That is, the MPU 10 accesses the data structure stored in the external storage device 18, and the data identification name recorded in the data identification name area (Name) of the data string matches the input data identification name. Data string (DS_
I. In the fourth embodiment, the data string with the ID number 6) is read from the data structure, and the contents of the input data string are stored in the variable X in the main storage device 14. In addition, M
The PU 10 accesses the data structure stored in the external storage device 18 and reads the data identifier name area (Nam
A data string (DS_P, data string of ID number 7 in the fourth embodiment) in which the data identification name recorded in e) matches the input parent concept name is read from the data structure, and the parent concept data is read. The contents of the column are stored in a variable Y in the main storage device 14. The structure of the variables X and Y can be the same as the structure of the data string. If the reading of the data string from the data structure has failed, the search processing ends (step-S73). When a plurality of parent concept names indicating a set including the data identification name are input, a plurality of variables Y may be set.

Next, it is checked whether or not link information on the parent concept data string (DS_P) is recorded in the parent data area of the input data string (DS_I). Specifically, M
The PU 10 stores the I stored in the variable X in the main memory 14.
The parent identification number area (Parent_ID_X) of the parent data area (Parent_X) in the content of the input data string of D number 6 is read, and the ID number 7 is read into the parent identification number area (Parent_ID_X).
Is checked (step-S74). If,
If it exists, the addition of link information is stopped because it violates the rule described above.

If no link information on the parent concept data string (DS_P) is recorded in the parent data area of the input data string (DS_I), the rule is checked. That is, a data string is extracted based on the link information recorded in the parent data area of the parent concept data string (DS_P) (ID number 7), and based on the link information recorded in the parent data area of the extracted data string, Data strings are extracted until the parent data area of the extracted data strings becomes empty (step-S7).
5). Specifically, the MPU 10 determines the parent identification number area (Parent_ID_) of the parent data area (Parent_Y) of the parent concept data string (DS_P) stored in the variable Y in the main storage device 14.
Based on the data of Y), the above-described search pattern 1 or search pattern 2 is executed. In this search step, data strings of ID number 12, ID number 15, and ID number 16 are extracted.

Then, one of the data strings extracted in this search step is the input data string (DS_I) (I
If the number matches the D number 6), the addition of the link information is stopped because the rule is violated (step-S76).

If none of the data strings extracted in this search step matches the input data string (DS_I), the parent concept data string (ID_6) is added to the parent data area of the input data string (DS_I) (ID number 6). The ID number 7, which is link information on DS_P), is recorded (step-S7).
7). That is, the MPU 10 stores the variable X in the main storage device 14.
The ID number 7 is stored in a new parent identification number area (Parent_ID_X) of the parent data area (Parent_X) in the content of the input data string (DS_I) with the ID number 6 stored in.

Next, the MPU 10 determines in step S78
The adjustment process of the data level value is executed. That is, as shown in FIG. 16, the MPU 10 outputs the input data sequence (DS_I)
(ID number 6) and the parent concept data sequence (D
S_P) is compared with the data level value. More specifically, the MPU 10 stores the data level value of the data level value area (Level_X) of the input data string stored in the variable X in the main storage device 14 and the data of the parent concept data string stored in the variable Y. The data level value in the level value area (Level_Y) is compared (step-S80). When the value of the data level value area (Level_Y) ≦ the value of the data level value area (Level_X), a value obtained by adding 1 to the value of the data level value area (Level_X) is stored in the data level value area (Level_Y). (Step-S81). In the fourth embodiment, the value of the data level value area (Level_Y) is equal to the value of the data level value area (Level_X).
Save.

Thereafter, the contents of the variables X and Y are written in the data string of the corresponding ID number in the data structure (step-S82). Thus, link information addition for adding new link information to the parent data area of the data string (DS_I) is completed.

By changing the data level value of the data string of ID number 7 from “1” to “2”, the data identification name indicating the set including the data identification name recorded in the data string of ID number 7 is changed. The data level value of the recorded data string (ID number 12 in the fourth embodiment) changes. Further, a data string recording a data identifier indicating a set including the data identifier recorded in the data string of ID number 12 (in the fourth embodiment, ID number 15 and ID number 16
The data level value of the data sequence changes. Therefore, readjustment processing of the data level value (step-S79) may be required.

In such a case, the data string is extracted based on the link information recorded in the parent data area in the data string (DS_P) in which the data identification name corresponding to the parent concept name is recorded, and the parent of the extracted data string is extracted. Based on the link information recorded in the data area, the data string is extracted until the parent data area of the extracted data string becomes empty. The data level value of the extracted data sequence (DS-A) is based on the link level recorded in the parent data area of the data sequence (DS-A). If it is equal to or greater than the value, a value obtained by adding 1 to the data level value of the extracted data string (DS-A) is added to the data string (DS-B) extracted based on the link information recorded in the parent data area.
In the data level value area.

For this purpose, as shown in FIG.
10 prepares the dynamic array F in the variable area and initializes the dynamic array F (Step-S90). This dynamic array F is
ID number area (ID_D) and data level value area (Level_
D). Then, the MPU 10 changes the ID number stored in the identification number area (ID_Y) and the data level value stored in the data level value area (Level_Y) of the data string stored in the variable Y in the main storage device 14 in a dynamic manner. Array F
ID number area (ID_D) and data level value area (Level_
D) (step-S91). In Example 4,
In this step-S91, I of the dynamic array F (0)
D number area (ID_D) and data level value area (Level_D)
Is stored with the ID number 7 and the data level value 2. Then, the variable i is set to 0 (step-S92).

Next, the MPU 10 reads out the data string corresponding to the ID number stored in the dynamic array F (i) from the data structure and stores it in the variable X (Step-S9).
3). Note that the structure of the variable X can be the same as the structure of the data string. In this step-S93, I
A data string corresponding to D number 7 is read from the data structure and stored in a variable X.

Next, the parent data area (Pare
nt_X), the parent identification number area (Parent_ID_X) is checked, and if the parent identification number area (Parent_ID_X) is empty, step-S
It moves to 101 (step-S94). If the parent identification number area (Parent_ID_X) is not empty, the MPU 10 determines the ID stored in the first parent identification number area (Parent_ID_X).
Read the data string corresponding to the number from the data structure,
The ID number and the data level value of the read data string are stored in variables P_ID and P_Level (step-
S95). In the fourth embodiment, in step S95,
I stored in the first parent identification number area (Parent_ID_X)
The data string corresponding to the D number 12 is read from the data structure, and the ID number 12 and the data level value “2” of the read data string are stored in the variables P_ID and P_Level.

Next, when the data level value stored in the variable P_Level is larger than the data level value stored in the data level value area (Level_D) of the dynamic array F (i) (step S96), Move to S99. On the other hand, the data level value stored in the variable P_Level is
When the data level is equal to or less than the data level value stored in the data level value area (Level_D) of the dynamic array F (i) (step-S9
6) The ID number stored in the variable P_ID and the value obtained by adding 1 to the data level value stored in the data level value area (Level_D) of the dynamic array F (i) are the ID number of the dynamic array F It is added to the area (ID_D) and the data level value area (Level_D) (step-S97). In the fourth embodiment, the data level value “2” stored in the variable P_Level is
Since the data level value is equal to the data level value stored in the data level value area (Level_D) of the dynamic array F (0), Step-S97
In the dynamic array F, the ID number 12 stored in the variable P_ID and the value “3” obtained by adding 1 to the data level stored in the data level value area (Level_D) of the dynamic array F (i) are added. Are added to the ID number area (ID_D) and the data level value area (Level_D) in the element at the end of. That is, the ID number 1 (ID_D) is assigned to the ID number area (ID_D) of the dynamic array F (1).
2 is stored, and the data level value “3” is stored in the data level value area (Level_D) of the dynamic array F (1).

Further, the MPU 10 stores the data level value area (Level_D) of the dynamic array F (i) in the data level value area (Level) of the data string of the data structure corresponding to the ID number stored in the variable P_ID. ) Is written with a value obtained by adding 1 to the stored data level value (step-S98). In the fourth embodiment, in this step-S98, the variable P_I
D in the data level value area (Level) of the data string of the data structure corresponding to the ID number 12 stored in the dynamic array F
A value “3” is obtained by adding 1 to the data level value stored in the data level value area (Level_D) of (i).

Next, the MPU 10 sets the parent identification number area (Parent_X) next to the parent data area (Parent_X) in the variable X.
_ID_X) (step-S99), and if the parent identification number area (Parent_ID_X) is not empty, the MPU 10 determines whether the I stored in the next parent identification number area (Parent_ID_X).
The data string corresponding to the D number is read from the data structure, and the ID number and data level value of the read data string are stored in the variables P_ID and P_Level (Step-S100), and the process returns to Step-S96.

The next parent identification number area (Parent_ID_
If X) is empty, the MPU 10 increments the value of i by one (step-S101), and checks whether data exists in the dynamic array F (i) (step-S102).
If data exists in the dynamic array F (i), the step
It returns to S93. If no data exists in the dynamic array F (i), the data level value readjustment process ends.

In the fourth embodiment, since the parent identification number area (Parent_ID_X) next to the variable X is empty, the value of i is set to 1
Is incremented by one (step-S101), and it is checked whether data exists in the dynamic array F (1) (step-S101).
102). Since data exists in the dynamic array F (1), the process returns to step S93.

Table 11 below shows the contents of the dynamic array F obtained up to the end of the data level value readjustment processing.
Table 12 shows the result of the readjustment processing of the data level value in the data structure shown in Table 3.

[0173]

[Table 11] F (i) P_ID P_Level F (0) 7 2 F (1) 12 3 F (2) 15 4 F (3) 16 4

[0174]

[Table 12]

In the fourth embodiment, one parent concept name indicating a set including a data identification name is input, but link information can be added by inputting a plurality of parent concept names.

Example 5 Example 5 relates to a data search method according to the second aspect of the present invention. In the fifth embodiment, a data string corresponding to the element concept and / or a data string corresponding to the indirect element concept are extracted from the data identifier to be searched.

The structure of the data string in the fifth embodiment is shown in FIG.
9 (A). The data string includes a data identifier name area (Name) in which a data identifier is recorded, a data level value area (Level) in which a data level value is recorded, and an element data area (Element). The data string further includes an identification number area (ID) in which an identification number for identifying the data string is recorded, and this identification number is a number unique to each data string.

In the data structure in the fifth embodiment, the data level value area (Level) of each data string includes:
It is assumed that the value of 1, 2, or 3 is recorded. That is,
The data structure in the fifth embodiment includes a plurality of first level data strings, a plurality of second level data strings, and a plurality of third level data strings.

In the element data area (Element) of each data string (DS_1), a data string (DS_0) recording a data identification name which is an element of a set represented by the data identification name recorded in the data string (DS_1) ) Is recorded. Element data area (Element)
Is composed of a plurality of element identification number areas (Element_ID). The data level value of the data string (DS_0) corresponding to the link information recorded in the element data area (Element) is smaller than the data level value of the data string (DS_1) including the element data area (Element).

The m-th level data identifier (where m = 2,
(3,..., N)
Level data identifier (however, m '= 1, 2, ..., m-
If 1) exists, the link information on the m'th level data string is stored in the element data area (Elemen
Recorded in t). Specifically, the identification number of the m'th level data string (DS_0) is recorded in the element identification number area (Element_ID) of the element data area (Element) in the mth level data string (DS_1). The identification number of the m'th level data string recorded in the element identification number area (Element_ID) of the element data area (Element) in the mth level data string is the link information in the fifth embodiment.

With reference to the m-th level data sequence, the m-th level data sequence corresponds to the element concept. Also, the m'th
The data identification name recorded in the level data sequence corresponds to the element concept name described above. On the other hand, when the m-th level data sequence is used as a reference, the m-th level data sequence corresponds to the parent concept. When the data identifier recorded in the m'th level data string is used as a reference, the data identifier recorded in the m'th level data string corresponds to the parent concept name. Further, the data identification name of the m'th level data string (DS_0) relating to the link information recorded in the element identification number area (Element_ID) of the element data area (Element) of the mth level data string is a set of element concepts. Equivalent to.

Table 3 shows a specific example of the data identifier corresponding to the concept name and its data level value, and the data identifier corresponding to the element concept name and its data level value in the data structure of the fifth embodiment. It is the same as the data structure shown.

A specific data string structure of the first level data string indicated by ID number 5 in Table 3 is exemplified in FIG. The data identification name in this first level data string is “insulator”, and this data identification name is recorded in the data identification name area (Name) of the data string. In the data level value area (Level_Value), “1” indicating that the data string is the first level data string is recorded.

Further, in the first level data sequence in which the first level data identifier "insulator" is recorded, there is no element of the set represented by the first level data identifier "insulator". Nothing is recorded in the element identification number area (Element_ID) of the data area (Element). The ID numbers 1 to 4 and 6 to 11, which are the first level data string, have the same structure.

[0185] In Table 3, the second (=
m) A specific data string structure of the level data string is illustrated in FIG. The second in this second level data sequence
The data identifier that is the level data identifier is “tree”,
This data identifier is the data identifier name area (Na
me). The data level value area (Leve
In “l_Value”, “2” indicating that the data string is the second level data string is recorded.

Further, the second-level data string in which the data identification name “tree” is recorded includes the m′-th level data identification name (in this case, m-th element) which is an element of the set represented by the data identification name “tree”. (= 1) exists (specifically, the data identifiers “insulator”, “flammable”, “plant” and “specific gravity 1”).
The following is a data string that records “)”, and the identification numbers 5 and 5 of the m′th level data string that record these m′th level data identifiers
6, 7, and 8 are recorded as link information in the element identification number area (Element_ID) of the element data area (Element) in the second level data string indicated by ID number 12.

ID numbers 13, 1 which are the second level data string
4 has a similar structure.

In Table 3, the third (=) indicated by ID number 15
N) A specific data string structure of the level data string is illustrated in FIG. The data identification name in this third level data string is “desk”, and this data identification name is recorded in the data identification name area (Name) of the data string. In the data level value area (Level_Value), “3” indicating that the data string is the third level data string is recorded.

In the third level data sequence in which the data identifier “desk” is recorded, the m′-th level data identifier (in this case, an element of the set represented by the data identifier “tree”)
m ′ = 1, 2) (specifically, a data string in which data identifiers “office supplies”, “furniture” and “tree” are recorded), and these m′-th level data identifiers are recorded. The identification numbers 1, 2, and 12 of the m'th level data string are recorded as link information in the element identification number area (Element_ID) of the element data area (Element) in the third level data string indicated by ID number 15. ing.

In Table 3, ID No. 16 and ID No. 1
The specific data sequence structures of the third level data sequence and the second level data sequence indicated by 4 are illustrated in FIGS. 21A and 21B, respectively.

To eliminate the cyclic definition, the data string (DS
_1) does not include link information (for example, an identification number) relating to the data string (DS_1) itself. Moreover, the data string (DS_
The link information recorded in the element data area of 1) is all different. These correspond to the rules and rules of [Construction rules of concept] described above.

Further, starting from any link information recorded in the element data area of the data string (DS_1), the link recorded in the element data area of the data string until the element data area of the data string becomes empty. When the data sequence is sequentially traced based on the information, the link information on the data sequence (DS_1) is not included in the element data area of any data sequence. This corresponds to the rule 'of the concept configuration rule' described above.

For example, the ID number 12 which is the link information recorded in the element identification number area (Element_ID) of the element data area (Element) of the third level data string (DS_1) indicated by the ID number 15 is changed to an arbitrary link. When information
Starting from this link information, the data sequence is sequentially traced based on the link information recorded in the element data area of the data sequence. That is, based on ID number 5 which is link information recorded in the element identification number area (Element_ID) of the element data area (Element) of the data string of ID number 12, the data string of ID number 5 is reached. The element identification number area (Element_I) of the element data area (Element) of the data string of ID number 5
D) is empty. Therefore, next, based on the ID number 6 which is the link information recorded in the element identification number area (Element_ID) next to the element data area (Element) of the data string of ID number 12, the data string of ID number 6 to arrive. ID
The element identification number area (Element_ID) of the element data area (Element) of the data string of No. 6 is empty. ID number 12
The same applies to the ID numbers 7 and 8 which are the link information recorded in the other element identification number areas (Element_ID) of the element data area (Element) of the data string. In such processing, the ID number 15 which is the link information on the data string (DS_1) is stored in the element data area of any of the data strings (in this example, the data strings of ID numbers 12, 5, 6, 7, and 8). Is not included.

When the third level data string indicated by ID number 15 is used as a reference, the second level data string of ID number 12 corresponds to the element concept, and the third level data string of ID numbers 5, 6, 7, and 8 One level data sequence corresponds to the concept of indirect elements. Conversely, ID
With reference to the first level data string indicated by No. 5,
The second level data string of ID number 12 corresponds to the parent concept, and the third level data string of ID number 15 corresponds to the indirect parent concept.

In the fifth embodiment, first, a data identifier to be searched is input. Next, the data string (DS_) in which the input data identification name is recorded in the data identification name area.
(1, input data string), extracting a data string based on the link information recorded in the element data area, and extracting the element data of the extracted data string based on the link information recorded in the element data area of the extracted data string. Data strings are extracted until the area becomes empty. The search pattern of the data search method according to the fifth embodiment is set to the search pattern 3 shown in FIG.

The data search method of the fifth embodiment will be described below with reference to FIG.
2 and the flowchart of FIG.

As shown in FIG. 22, first, a character string "desk", which is a data identification name corresponding to data to be searched, is input from the input device 12 composed of, for example, a keyboard (step-S110). The MPU 10 reads the input character string “desk” and stores it in a variable area (not shown) in the main storage device 14. Next, the MPU 10 accesses the data structure stored in the external storage device 18 and stores the data structure in which the data identification name recorded in the data identification name area (Name) of the data string matches the input character string. A column (which is an input data column, and in the fifth embodiment, a data column of ID number 15) is read from the data structure (step-S
111), the contents of the input data string are stored in the main storage device 1
4 is stored in the variable area. If the reading of the data string from the data structure has failed, the search process ends (step-S112).

Next, the MPU 10 checks the element identification number area of the element data area of the input data string stored in the variable area in the main storage device 14, and terminates the search processing if the element identification number area is empty. (Step-S113). If the element identification number area is not empty, the MPU 10 prepares a dynamic array D (0) in the variable area and initializes the dynamic array D (0) (Step-S114). Note that the dynamic array D stores the ID numbers of the data strings. After that, the ID number of the input data string is stored in the dynamic array D (0) (Step-S115).

Next, a data string is extracted based on the link information recorded in the element data area of the input data string, and the extracted data string is extracted based on the link information recorded in the element data area of the extracted data string. Data strings are extracted until the element data area becomes empty. Specifically, search processing 3 shown in FIG. 23 is executed (step-S116). In the search process 3, first, the variable i is set to 0 (step-S120).

[0200] Then, the MPU 10 has the dynamic sequence D (i).
Is read from the data structure and stored in the variable X (step-S12).
1). Note that the structure of the variable X can be the same as the structure of the data string. Next, the element identification number area (Element_ID_X) of the element data area (Element_X) in the variable X is checked, and if the element identification number area (Element_ID_X) is empty,
It moves to step-S128 (step-S122). If the element identification number area (Element_ID_X) is not empty, the I stored in the first element identification number area (Element_ID_X)
The D number is stored in the variable EID (Step-S12)
3).

Next, the MPU 10 checks whether or not the ID number stored in the variable EID exists in the dynamic array D (Step-S124).
Move to 126. On the other hand, if it does not exist, the variable E
The ID number stored in the ID is added to the end of the dynamic array D (step-S125). Next, the MPU 10 checks the element identification number area (Element_ID_X) next to the element data area (Element_X) in the variable X, and if this element identification number area (Element_ID_X) is empty, Step-S128
(Step-S126). Element identification number area (El
If ement_ID_X) is not empty, the ID number stored in this element identification number area (Element_ID_X) is stored in a variable EID (step-S127), and then step-S12
Return to 4.

In step S128, the value of the variable i is incremented by one, and then the dynamic array D (i)
Is stored in the dynamic array D
It is checked whether data exists in (i) and the dynamic array D (i)
If the ID number is stored in step S121, the process returns to step S121. If the ID number is not stored in the dynamic array D (i), the search process 3 ends.

Thereafter, the MPU 10 executes the dynamic array D (i)
The data string corresponding to the ID number stored in the data structure is read from the data structure, and the data identification name recorded in the data identification name area and the data level value recorded in the data level value area of the data string are displayed, for example, on the display device 16. (Step-S117). The data identifier to be searched may or may not be displayed. Thus, the search processing ends.

When "desk" is searched, the following Table 13 is used.
The ID numbers are stored in the dynamic array D (i) in the order shown in FIG.

[0205]

[Table 13] D (0): ID number = 15 D (1): ID number = 1 D (2): ID number = 2 D (3): ID number = 12 D (4): ID number = 5 D (4): ID number = 6 D (4): ID number = 7 D (4): ID number = 8

Thus, in the fifth embodiment, when "desk" is input as the data identifier to be searched, the first level data string in which the first level data identifier which is an element of the set represented by "desk" is recorded. Not only are "office supplies" and "furniture", and "trees", which are second-level data strings that record second-level data identifiers, extracted, but also "trees".
It is possible to extract "insulator", "flammable", "plant", and "specific gravity 1 or less", which are the first-level data strings that record the first-level data identifiers that are elements of the set represented by. That is, even if an element such as “insulator” is not input to “desk”, “insulator”, “flammable”, “plant” and “specific gravity 1 or less” are used as “indirect element concepts” for “desk”. Is included. The search results are shown in Table 14 below. For reference, after the data level value, whether it is an element concept or an indirect element concept is shown.

[0207]

[Table 14] Data level value of the extracted data string Concept classification Data identification name Office supplies 1 Element concept Furniture 1 Element concept Tree 2 Element concept Insulator 1 Indirect element concept Flammability 1 Indirect element concept Plant 1 Indirect element concept Specific gravity 1 or less 1 Indirect element concept

In the fifth embodiment, the third level data identifier is searched for. However, depending on the value of N, the second to Nth level data identifiers are searched. You can also. Although the search pattern 3 is employed in the fifth embodiment, the search pattern 4 can be used instead.

(Embodiment 6) Embodiment 6 also relates to a data search method according to the second aspect of the present invention. In the fifth embodiment, one data identifier to be searched is set. On the other hand, in the sixth embodiment, there are a plurality of data identifiers to be searched. Then, the data search method described in the fifth embodiment is executed for each data identifier to be searched. Then, for each of the extracted data columns,
Outputs the number of related data identifiers to be searched. That is, the sixth embodiment relates to a scrap-type data search method.

Specifically, the sixth embodiment will be described by taking the data structure described in the fifth embodiment as an example. Although the search pattern is the same as the search pattern 3 in the fifth embodiment, the search pattern 4 can also be used. There were two data identifiers to be searched, namely "desk" and "pencil". still,
The data level values of the data strings in which the data identifiers to be searched are recorded may be the same or different. Example 6
Hereinafter, the data search method will be described with reference to the flowcharts of FIGS.

As shown in FIG. 24, first, the character strings "desk" and "pencil" are input from the input device 12 composed of, for example, a keyboard (step S130). MPU10
Reads the input strings "desk" and "pencil"
It is stored in a variable area (not shown) in the main storage device 14.
Next, the MPU 10 substitutes “1” for the variable j,
The data structure stored in the external storage device 18 is accessed, and the data identification name recorded in the data identification name area (Name) of the data string (in the sixth embodiment, the data string of ID number 15) is input. A data string (first input data string) that matches the input character string is read from the data structure, and the contents of the data string are stored in a variable area in the main storage device 14. This process is executed by the number of input data identifiers. That is, in the sixth embodiment,
The MPU 10 accesses the data structure stored in the external storage device 18, reads the data string with the ID number 16 (the second input data string) from the data structure, and stores the contents of the data string in the main storage. It is stored in a variable area in the device 14 (step-S131). If the reading of the data string from the data structure has failed, the search process ends (step-S132). Next, the MPU 10 checks the element identification number areas of the element data areas of all the data strings stored in the variable areas in the main storage device 14. If the element identification number areas of all the data strings are empty, the MPU 10 performs a search process. Is ended (step-S133).

On the other hand, if the element identification number area of any of the data strings is not empty, MPU 10 prepares dynamic array D (0) in the variable area and initializes dynamic array D (0) ( Step-S134). Note that the dynamic array D stores the ID numbers of the data strings. After that, among the data strings stored in the variable area in the main storage device 14, the ID number 15 of the first input data string which is the data string in which the first data identifier to be searched is recorded is assigned to the dynamic array D. (0) (step-S135).

Next, a data string is extracted based on the link information recorded in the element data area of the first input data string, and extracted based on the link information recorded in the element data area of the extracted data string. The data string is extracted until the element data area of the set data string becomes empty. Specifically, a search process 3 similar to that of the fifth embodiment is executed (step-S136).

Thereafter, the MPU 10 reads the sequence D _j (= D ₁ )
Is prepared, the contents of the dynamic array D (i) are copied to the array D _j (i), and the dynamic array D (i) is discarded (step-S13).
7). Next, the MPU 10 checks whether or not the next input data string is stored in the variable area in the main storage device 14 (Step-S138). If the next input data string is not stored in the variable area in the main storage device 14, the process proceeds to step S50 shown in FIG. Incidentally, the content of the D _1, shown in the following Table 15.

[0215]

[Table 15] D ₁ (0): ID number = 15 D ₁ (1): ID number = 1 D ₁ (2): ID number = 2 D ₁ (3): ID number = 12 D ₁ (4): ID number = 5 D ₁ (5): ID number = 6 D ₁ (6): ID number = 7 D ₁ (7): ID number = 8

If the next input data string is stored in the variable area in the main storage device 14, the variable j is incremented by one, and the process returns to step S134. That is, M
The PU 10 prepares the dynamic array D (0) in the variable area, and initializes the dynamic array D (0) (Step-S134).
After that, the ID number 16 of the second input data string is stored in the dynamic array D (0) (Step-S135).

Next, a data string is extracted based on the link information recorded in the element data area of the second input data string, and extracted based on the link information recorded in the element data area of the extracted data string. The data string is extracted until the element data area of the data string becomes empty. Specifically, a search process 3 similar to that of the fifth embodiment is executed (step-S136).

Thereafter, the MPU 10 reads the sequence D _j (= D ₂ )
Is prepared, and the contents of the dynamic array D (i) are copied to the array D ₂ ,
The dynamic array D (i) is discarded (step-S137).
Incidentally, the content of D _2, shown in Table 16 below.

[0219]

[Table 16] D ₂ (0): ID number = 16 D ₂ (1): ID number = 1 D ₂ (2): ID number = 3 D ₂ (3): ID number = 4 D ₂ (4): ID number = 12 D ₂ (5): ID number = 13 D ₂ (6): ID number = 5 D ₂ (7): ID number = 6 D ₂ (8): ID number = 7 D ₂ (9): ID number = 8 D ₂ (10): ID number = 9 D ₂ (11): ID number = 10 D ₂ (12): ID number = 11

Next, the MPU 10 checks whether or not the next input data string is stored in the variable area in the main storage device 14 (step S138). In the sixth embodiment, since the next data string is not stored in the variable area in the main storage device 14, the process proceeds to step S50 (see FIG. 14). The processing in step-S50 to step-S60 is the same as that in the third embodiment, and thus detailed description is omitted. Table 17 below shows a display example after the processing is completed in these steps. For reference, the input data identifier to be searched is attached after the number of hit input data strings. In Table 17, the number of hit input data strings is sorted in descending order.

[0221]

[Table 17] Extracted data Number of input data rows that were hit Data identifiers of columns to be searched Data identifiers Office supplies 2 desks / pencils Trees 2 desks / pencil insulators 2 desks / pencils Flammable 2 desks / pencils Plants 2 desk / pencil Specific gravity 1 or less 2 desk / pencil furniture 1 desk writing implement 1 pencil columnar 1 pencil graphite 1 pencil conductivity 1 pencil black 1 pencil soft 1 pencil

In the sixth embodiment, the number of data identifiers to be searched, which are related to each of the extracted first and second level data strings, is output.
It should be noted that the data level value to be searched, which is related to the extracted data sequence having the data level value, is determined in advance by inputting the data level value from the input device 12. It should be left.

In such a scrap method, the value recorded in the count area (Count_R) of the dynamic array R (the number of hit input data strings) is “desk” and “desktop” which are data identification names to be searched. The two data identifiers "pencil" indicate how many contain or relate to the extracted data string. For example, since the number in the number field in the first level data string having “insulator” as the data identifier is “2”, the two data identifiers to be searched are the first level concept name “insulator” Is included directly or indirectly.

Here, directly containing means that the concept represented by the data identifier to be searched is the parent concept of the extracted concept. On the other hand, including indirectly means that the concept represented by the data identifier to be searched is an indirect parent concept of the extracted concept. Specifically, the third
The level concept name “desk” directly includes the first level concept name “office supplies”. On the other hand, the third level concept name “desk” indirectly includes the first level concept name “insulator”.

Here, the extracted data string that directly or indirectly contains all of the data identifiers to be searched is the search data obtained when the data identifiers to be searched are connected by the conventional search condition “AND”. Corresponds to the result. Therefore, Example 6
In, since two data identifiers were searched,
Whether the concept to be searched or the indirect identification name common to the concepts to be searched can be easily known by reading the data identification name in which the number of hit input data strings is “2”.

An extracted data string including at least one data identifier to be searched for directly or indirectly corresponds to a search result obtained when the conventional search condition “OR” is used. Here, in the sixth embodiment, since two data identifiers are searched, it is determined whether an element concept or an indirect identifier not common to the two concepts to be searched is the number of hit input data strings. By reading the data identification name of "1", it can be easily known.
As described above, unlike the conventional data search method, it is possible to easily check how many data identifiers to be searched are related to each of the extracted data strings.

In the sixth embodiment, the data identifier recorded in the third level data string is searched. However, depending on the value of the level N, the search target is the second level data string to the second level data string. The data identifiers recorded in the N-level data string may be mixed.

(Embodiment 7) Embodiment 7 relates to a link information adding method according to the second aspect of the present invention. The structure of the data string in the data structure of the seventh embodiment and various data recorded in each data string were the same as in the fifth embodiment. Less than,
Embodiment 7 with reference to the flowcharts of FIGS.
Will be described.

In the seventh embodiment, a new data string having the ID number 18 and the data identification name “motor” is generated in advance as the first level data string, and this data string is stored in the external storage device 18. Suppose it is added to the structure.

As shown in FIG. 25, first, a data identification name is inputted from the input device 12 comprising, for example, a keyboard, and at the same time, an element concept name which is an element of a set represented by this data identification name is inputted ( Step-S140).
In the seventh embodiment, the data identification name is “motor”,
The concept name of the element is "motor brush". MPU10
Reads the input character strings “motor” and “motor brush” and stores them in a variable area set in the main storage device 14. Then, the MPU 10 checks whether or not the input data identification name matches the input element concept name (step-S141). If the data identification name matches, the MPU 10 violates the rule described above, and stops adding the link information. .

If the input data identification name does not match the input element concept name, a data string (input data string, DS_I) recording the input data identification name is extracted. That is, the MPU 10 accesses the data structure stored in the external storage device 18, and the data identification name recorded in the data identification name area (Name) of the data string matches the input data identification name. The data sequence (DS_I) (in the seventh embodiment, the data sequence of ID number 18) is read from the data structure (step-S142), and the contents of the input data sequence are stored in the variable X in the main storage device 14. . At the same time, the MPU 10 accesses the data structure stored in the external storage device 18, and matches the data identification name recorded in the data identification name area (Name) of the data string with the input element concept name. Data string (DS_E)
(In the seventh embodiment, the data string of ID number 17) is read from the data structure (step-S142), and the contents of the data string (sometimes called an element concept data string) are stored in the main storage device 14. Store in variable Y. Variable X,
The structure of Y can be the same as the structure of the data string. If the reading of the data string from the data structure has failed, the search processing ends (step-S143). In this way, a data string in which the input data identification name is recorded can be extracted, and a data string in which the data identification name that matches the input element concept name can be extracted.
When a plurality of element concept names, which are elements of a set represented by a data identification name, are input, a plurality of variables Y may be set.

Next, it is checked whether or not link information relating to the element concept data string (DS_E) is recorded in the element data area of the input data string (DS_I). Specifically, the MPU 10 reads the element identification number area (Element_ID_X) of the element data area (Element_X) in the contents of the input data string (DS_I) of the ID number 18 stored in the variable X in the main storage device 14, Element identification number area (El
ement_ID_X) is checked for an ID number 17 (step-S144). If there is, the addition of link information is stopped because it violates the rules described above.

If no link information relating to the element concept data string (DS_E) is recorded in the element data area of the input data string (DS_I), the rule 'is checked. That is, the element concept data string (DS_E) (ID number 1
The data string is extracted based on the link information recorded in the element data area of 7), and the element data area of the extracted data string is emptied based on the link information recorded in the element data area of the extracted data string. Until the data string is extracted (step-S145). Specifically, the MPU 10
Based on the data of the element identification number area (Element_ID_Y) of the element data area (Element_Y) of the element concept data string stored in the variable Y in the main storage device 14, the search pattern 3 or the search pattern 4 described above is executed. . In this search step, data strings of ID number 13, ID number 9, ID number 10, and ID number 11 are extracted.

Then, one of the data strings extracted in this search step is the input data string (DS_I) (I
D number 18), it is against the rule '
The addition of the link information is stopped (step-S146).

If none of the data strings extracted in this search step matches the input data string (DS_I), the element concept data string (ID_18) is stored in the element data area of the input data string (DS_I) (ID No. 18). The ID number 17 which is link information regarding DS_E) is recorded (step-S147). That is, the MPU 10 stores the ID number 17 in a new parent identification number area (Element_ID_X) of the element data area (Element_X) in the content of the input data string of the ID number 18 stored in the variable X in the main storage device 14. I do.

Next, the MPU 10 executes Step-S14
8 is performed. That is, FIG.
As shown in FIG. 3, the MPU 10 receives the input data sequence (DS_
I) The data level value of (ID number 18) is compared with the data level value of the element concept data sequence (DS_E). More specifically, the MPU 10 controls the data level value area (Level_Level) of the input data string stored in the variable X in the main storage device 14.
The value of (X) is compared with the value of the data level value area (Level_Y) of the element concept data string stored in the variable Y (step-S150). Then, the data level value area (Level_
If the value of X) ≦ the value of the data level value area (Level_Y), a value obtained by adding 1 to the value of the data level value area (Level_Y) is stored in the data level value area (Level_X) (step-S).
151). In the seventh embodiment, the data level value area (Le
vel_X) (= 1) <data level value area (Level_Y) value (= 2), the data level value area (Level_X)
Save "3" in

Thereafter, the contents of the variables X and Y are written in the data string of the corresponding ID number in the data structure (step-S152). Thus, link information addition for adding new link information to the element data area of the data string is completed.

By changing the data level value of the data string of ID number 18 from “1” to “3”,
The data level value of the data string in which the data identification name indicating the set including the data identification name recorded in the data string is also changed. Therefore, readjustment processing of the data level value (step-S149) may be required. In the seventh embodiment, there is no such data string.

In such a case, all data strings having a data level value larger than the data level value before adjustment in the input data string are extracted. Then, for each of the extracted data strings (referred to as a departure data string), the data string is extracted based on the link information recorded in the element data area in the data string according to the search pattern 3 or the search pattern 4, and extracted. Based on the link information recorded in the element data area of the extracted data string, until the element data area of the extracted data string becomes empty, or of the data string from which the link information related to the input data string (DS_I) is extracted. Data strings are extracted until found in the element data area. Then, when link information relating to the input data sequence (DS_I) exists in the extracted data sequence, the departure data sequence and a series of data sequences extracted from the departure data sequence are stored. On the other hand, if there is no link information on the input data sequence (DS_I) in the extracted data sequence, the departure data sequence is discarded.

Next, in each of the stored departure data strings, the data level values of the departure data string and a series of data strings extracted from the departure data string are adjusted in the input data string (DS_I). Re-adjust based on the data level value. That is, the data level value of a certain data sequence (DS-C) is the level value of the data sequence (DS-D) extracted based on the link information recorded in the element data area of the data sequence (DS-C). Is smaller than the maximum value, the value obtained by adding 1 to the maximum value of the level values of the extracted data sequence (DS-D) is recorded in the data level value of a certain data sequence (DS-C). I do. Such an operation is performed from the last data string of the series of data strings extracted based on the starting data string to the first data string, and further, to the starting data string.
Just do it.

In the seventh embodiment, one element concept name, which is an element of a set represented by a data identification name, is input. However, link information can be added by inputting a plurality of element concept names. it can.

Example 8 Example 8 relates to a data search method according to the third aspect of the present invention. In the eighth embodiment, a data sequence corresponding to a parent concept, a data sequence corresponding to an indirect parent concept, a data sequence corresponding to an element concept, and a data sequence corresponding to an indirect element concept are provided for a data identifier to be searched. Extract at least one. Note that the data structure in Example 8 was the same as in Table 12.

FIG. 27 shows an example of the structure of the data string of each level in the eighth embodiment. The data string in the eighth embodiment differs from the data strings in the first and fifth embodiments in that it has a parent data area (Parent) and an element data area (Element). That is, the data string in the eighth embodiment includes an identification number area (ID), a data identification name area (Name),
Data level value area (Level), parent data area (Parent)
And an element data area (Element).
The parent data area (Parent) includes a plurality of parent identification number areas (Pa
rent_ID) and element data area (Eleme
nt) is composed of a plurality of element identification number areas (Element_ID). In Example 8, N = 4.

The parent data area (Parent) of each data string (DS_1) relates to a data string (DS_2) recording a data identification name indicating a set including the data identification name recorded in the data string (DS_1). Link information is recorded. The parent data area (Parent) is composed of a plurality of parent identification number areas (Parent_ID). The data level value of the data string (DS_2) corresponding to the link information recorded in the parent data area (Parent) is larger than the data level value of the data string (DS_1) including the parent data area (Parent).

The M-th level data identifier (however, M = 1,
, M−1 level data identifiers (M ′ = M + 1, M + 2,.
N), the M'th level data string (DS_2)
Link information is recorded in the parent data area (Parent) of the M-th level data string (DS_1). Specifically, the identification number of the M'th level data string (DS_2) is recorded in the parent identification number area (Parent_ID) of the parent data area (Parent) in the Mth level data string (DS_1). Parent data area (Paren
M) recorded in the parent identification number area (Parent_ID) of t)
The identification number of the level data string is the link information recorded in the parent data area in the eighth embodiment.

As described in the first embodiment, in order to eliminate the cyclic definition, the parent data area (Parent) of the data string (DS_1) has link information (eg, identification number) related to the data string (DS_1) itself. ) Is not included.
In addition, the link information recorded in the parent data area of the data string (DS_1) is all different. These correspond to the rules and rules of [Construction rules of concept] described above.

Further, starting from any link information recorded in the parent data area of the data string (DS_1), the link recorded in the parent data area of the data string until the parent data area of the data string becomes empty. When the data sequence is sequentially traced based on the information, the link information on the data sequence (DS_1) is not included in the parent data area of any data sequence.
This corresponds to the rule of [Construction rules of concept] described above.

[0248] In the element data area (Element) of each data string (DS_1), a data string (DS_0) recording a data identification name that is an element of a set represented by the data identification name recorded in the data string (DS_1). ) Is recorded. The element data area (Elemen
data string (DS) corresponding to the link information recorded in (t)
_0) is set in the element data area (Elemen
t) is smaller than the data level value of the data string (DS_1) containing the data string.

The m-th level data identifier (where m = 2,
(3,..., N)
Level data identifier (however, m '= 1, 2, ..., m-
If 1) exists, the link information on the m'th level data string is stored in the element data area (Elemen
Recorded in t). Specifically, the identification number of the m'th level data string is recorded in the element identification number area (Element_ID) of the element data area (Element) in the mth level data string. The identification number of the m'th level data string recorded in the element identification number area (Element_ID) of the element data area (Element) in the mth level data string is the eighth embodiment.
Is the link information recorded in the element data area.

As described in the fifth embodiment, in order to eliminate the cyclic definition, an element data area (Element)
Contains link information about the data string itself (for example,
Identification number) is not included. Moreover, the link information recorded in the element data area of the data string is all different. These correspond to the rules and rules of [Construction rules of concept] described above.

Further, starting from arbitrary link information recorded in the element data area of the data string (DS_1), the link recorded in the element data area of the data string until the element data area of the data string becomes empty. When the data sequence is sequentially traced based on the information, the link information on the data sequence (DS_1) is not included in the element data area of any data sequence. This corresponds to the rule 'of the concept configuration rule' described above.

Embodiment 8 will be described below with reference to the data structures listed in Table 12.

FIG. 28 exemplifies a specific data string structure of the first level data string indicated by ID number 5 in Table 12.
The first level data identifier in the first level data string is "insulator", and this data identifier is recorded in the data identifier area (Name) of the data string. In the data level value area (Level), “1” indicating that the data string is the first level data string is recorded.

Level M ′ indicating the set including the first level data identifier “insulator” (in the eighth embodiment, M ′ =
2 or 3) Since the data identifier exists (specifically,
Second and third level data identifiers "tree" and "mica"),
The identification numbers 12 and 14 (corresponding to the link information relating to the parent concept) of the M′-th level data string in which these M′-level data identification names are recorded are the parent numbers in the first level data string indicated by ID number 5. It is recorded in the parent identification number area (Parent_ID) of the data area (Parent). In addition, ID number 5
Is a first-level data string, so the element identification number area (Element) of the element data area (Element)
_ID) has nothing recorded. The ID numbers 1 to 4 and 6 to 11, which are the first level data string, have the same structure.
In some cases, there is an Mth level data string in which there is no M'th level data identifier indicating a set including the Mth level data identifier. In this case, nothing is recorded in the parent identification number area (Parent_ID) of the parent data area (Parent) in the M-th level data string.

In Table 12, the third item indicated by ID number 12
FIG. 29 shows a specific data string structure of the (= M) level data string.
An example is shown below. The third level data identifier in the third level data string is “tree”, and this data identifier is recorded in the data identifier name area (Name) of the data string. In the data level value area (Level), the data string is
“3” indicating the level data string is recorded.

Further, the third-level data sequence in which the third-level data identifier “tree” is recorded includes the M′-level data identifier (implementation) indicating a set including the third-level data identifier “tree”. In Example 8, since M ′ = 4) exists (specifically, the fourth-level data identifiers “desk” and “pencil”), the M′-th level data identifiers of these M′-level data identifiers are recorded. The identification numbers 15 and 16 (corresponding to the link information related to the parent concept) of the level data string are the parent data area (Parent) in the third level data string indicated by the ID number 12.
Is recorded in the parent identification number area (Parent_ID).

The third-level data sequence in which the third-level data identifier is recorded further includes first and second (= m ')-level data identifiers which are elements of the set represented by the third-level data identifier. Link information on the first and second level data strings in which names are recorded is recorded in the element data area (Element) of the data string. Specifically, the identification numbers 5, 6, and 8 of the first level data sequence recording the first level data identifier and the identification number 7 (the element concept of the second level data sequence recording the second level data identifier) (Equivalent to related link information)
Is the element identification number area (Eleme) of the element data area (Element) in the third level data string indicated by ID number 12.
nt_ID).

ID number 1 which is the second and third level data strings
3, 14, and 17 also have the same structure.

In Table 12, the fourth item indicated by ID number 15
(= N) FIG. 30 shows a specific data string structure of the level data string.
An example is shown below. The fourth level data identifier in the fourth level data string is “desk”, and this data identifier is recorded in the data identifier name area (Name) of the data string. In the data level value area (Level), the data string is the fourth.
“4” indicating the level data string is recorded.

Further, the fourth (= N) th level data identifier recorded in the fourth level data string includes the third (= N-1) element which is an element of the set represented by the fourth level data identifier. Link information relating to the third level data string in which the level data identification name is recorded is recorded in the element data area (Element), and further, the L ′ level data identification which is an element of the set represented by the fourth level data identification name Name (however, L '= 1,
,..., N−1, and in the data string of ID number 15, link information on the L′ th level data string in which L ′ = 1, 2) is recorded in the element data area (Element). . More specifically, the identification number 12 (corresponding to link information relating to the element concept) of the third level data sequence in which the third level data identifier is recorded, and the first level in which the first level data identifier is recorded The identification numbers 1 and 2 of the data string (corresponding to link information related to the element concept) are stored in the element identification number area (Element_ID) of the element data area (Element) in the fourth level data string indicated by ID number 15. Has been recorded.

ID number 16, which is the fourth level data string, has a similar structure. In the eighth embodiment, since the maximum level of the data string is 4, the parent identification number area (Parent_I) of the parent data area (Parent) in the fourth level data string is used.
Nothing is recorded in D).

In Table 12, ID number 14 and ID number 16
The specific data string structures of the second-level data string and the fourth-level data string indicated by are exemplified in FIGS. 31 and 32, respectively.

In the search method of the eighth embodiment, first, a data identifier to be searched is input. Next, the input data identifier is extracted from the data string based on the link information recorded in the parent data area of the data string recorded in the data identifier name area, and is recorded in the parent data area of the extracted data string. Based on the link information, data strings are extracted until the parent data area of the extracted data string becomes empty. Such a search can be performed using the search pattern 1 shown in FIG. 1 and described in the first embodiment, or the search pattern 2 shown in FIG. 2 and described in the second embodiment. At the same time, the input data identification name is extracted from the data string based on the link information recorded in the element data area of the data string recorded in the data identification name area, and is recorded in the element data area of the extracted data string. Based on the link information, the data string is extracted until the element data area of the extracted data string becomes empty. Such a search can be performed using the search pattern 3 shown in FIG. 3 and described in the fifth embodiment, or the search pattern 4 shown in FIG. Then, the extracted data string is output. The search method has already been described in Embodiment 1, Embodiment 2,
Since it can be similar to the fifth embodiment, detailed description is omitted.

For example, when the data to be searched is “tree”, the search results shown in Table 18 below can be obtained.

[0265]

[Table 18] Data level value of the extracted data sequence Concept classification Data identification name Desk 4 Parent concept Pencil 4 Parent concept Insulator 1 Element concept Flammability 1 Element concept Plant 2 Element concept Specific gravity 1 or less 1 Element concept

Note that “flammable” is an element concept of “plant” and therefore corresponds to an indirect element concept of “tree”. Does not appear in the results.

For example, when the data to be searched is “plant”, the search results shown in Table 19 below can be obtained.

[0268]

[Table 19] Data level value of extracted data sequence Concept classification Data identification name Tree 3 Parent concept Desk 4 Indirect parent concept Pencil 4 Indirect parent concept Flammability 1 Element concept

In the eighth embodiment, the data search method can be performed by the scrap method described in the third or sixth embodiment. For example, when the data to be searched are “tree” and “mica”, search results shown in Table 20 below can be obtained.

[0270]

[Table 20] Number of input data strings that were hit in the extracted data strings Data identifier Name 1 Pencil 1 Insulator 2 Flammable 1 Plant 1 Specific gravity 1 or less 1

(Embodiment 9) Embodiment 9 relates to a data search method according to the fourth aspect of the present invention. The structure of the data string in the data structure of the ninth embodiment and various data recorded in each data string were the same as in the eighth embodiment.

In the search method of the ninth embodiment, first, a data identifier to be searched is input. Next, the input data identifier is extracted from the data string based on the link information recorded in the element data area of the data string recorded in the data identifier name area, and is recorded in the element data area of the extracted data string. Based on the link information, the data string is extracted until the element data area of the extracted data string becomes empty. Such a search can be performed using the search pattern 3 shown in FIG. 3 and described in the fifth embodiment, or the search pattern 4 shown in FIG.

Then, of the data strings extracted in this step, a data string is extracted based on the link information recorded in the parent data area of the data string whose element data area is empty, and the parent of the extracted data string is extracted. Based on the link information recorded in the data area, the data string is extracted until the parent data area of the extracted data string becomes empty. Then, the extracted data string is output. Such a search can be performed using the search pattern 1 shown in FIG. 1 and described in the first embodiment, or the search pattern 2 shown in FIG. 2 and described in the second embodiment.

Specifically, in order to evaluate whether or not the element data area is an empty data string, in addition to storing the ID number in the dynamic array D described in the fifth embodiment, the data level value is also determined. The stored dynamic sequence D ′ may be used (see FIGS. 22 and 23). Then, after completing the search processing 3 described in the fifth embodiment, if a new dynamic array D ′ including the dynamic array D (0) and the data level value whose data level value is “1” is generated, Good. Then, using the data identification name corresponding to the ID number recorded in the new dynamic array D ′ as the data identification name to be searched, for example, the search processing 1 described in the first embodiment (see FIG. 9) Execute. Then, as in step S17 of the first embodiment, a data string corresponding to the ID number stored in the dynamic array D ′ (i) is read from the data structure and recorded in the data identification name area of the data string. The data identification name and the data level value recorded in the data level value area may be displayed on the display device 16, for example.

For example, when the data to be searched is “tree”, a new dynamic array D ′ having the contents shown in Table 21 below is obtained.
(I) can be obtained. In addition, search results shown in Table 22 below can be obtained. In this case, the data level values were output in ascending order.

[0276]

D '(0): ID number = 12 D' (1): ID number = 5 D '(2): ID number = 6 D' (3): ID number = 8

[0277]

[Table 22] Data level value of the extracted data string Data identification name Insulator 1 Flammability 1 Specific gravity 1 or less 1 Plant 2 Mica 2 Tree 3 Machine 4 Pencil 4

In Example 9, a new dynamic sequence D ′
Based on the above, the data search method can be performed by the scrap method described in the third embodiment. In this case, a data string based on the ID number that is the content of the new dynamic array D ′ is sequentially read from the data structure, and the read data string is stored in the variable area as an input data string. It is sufficient to execute the processing from step S43 in step 3 onward.

Embodiment 10 Embodiment 10 relates to a link information adding method according to the third aspect of the present invention, and relates to a link information adding method for adding new link information to a parent data area of a data string. The structure of the data string in the data structure of the tenth embodiment and various data recorded in each data string were the same as in the first embodiment. The link information adding method according to the tenth embodiment is basically similar to the first link information adding method according to the present invention described in the fourth embodiment. Therefore, focusing on differences from the first link information adding method of the present invention described in the fourth embodiment, referring again to the flowcharts of FIGS. The adding method will be described.

As shown in FIG. 15, first, a data identification name is input from the input device 12 composed of, for example, a keyboard, and a parent concept name indicating a set including the data identification name is input (step-). S70). In the tenth embodiment, as new link information, link information relating to the data string of ID number 7 in which the data identification name “plant” is recorded is added to the data string of ID number 6 in which the data identification name “flammable” is recorded. . That is, the concept name “plant” corresponds to the parent concept, and the concept name “flammable” corresponds to the element concept.

The MPU 10 reads the input character strings “flammable” and “plant” and stores them in a variable area set in the main storage device 14. Then, the MPU 10 checks whether the input data identification name matches the input parent concept name (Step-S71). If the data identification name matches, the MPU 10 violates the rule described above, and stops adding the link information. . If the input data identification name does not match the input parent concept name, an input data sequence (DS_I) (ID number 6), which is a data sequence in which the input data identification name is recorded, is extracted. Parent concept data string (DS), which is a data string that records a data identifier that matches the parent concept name
_P) (ID number 7) (step-S72,
Step-S73).

Next, it is checked whether or not link information on the parent concept data string (DS_P) exists in the parent data area of the input data string (DS_I) (step-S74).
If it exists, the addition of the link information is stopped because it is contrary to the rule described above. Such a determination can also be made by determining whether or not link information relating to the input data string is recorded in the element data area of the parent concept data string.

If there is no link information relating to the parent concept data string (DS_P) in the parent data area of the input data string (DS_I), the rule is checked. That is, a data string is extracted based on the link information recorded in the parent data area of the parent concept data string (DS_P), and the extracted data string is extracted based on the link information recorded in the parent data area of the extracted data string. Until the parent data area becomes empty (step-S75). In this search step, data strings of ID number 12, ID number 15, and ID number 16 are extracted.

Then, one of the data strings extracted in this search step is the input data string (DS_I) (ID
If the number matches (No. 6), the rule is violated, and the addition of link information is stopped (step-S76).

Each of the data strings extracted in this search step is an input data string (DS_I) (ID number 6).
If not, the parent concept data string (DS_P) is stored in the parent data area of the input data string (DS_I) (ID number 6).
The ID number 7 which is link information on the ID is recorded (step-S77). In addition, parent concept data string (DS_P)
In the element data area in (ID number 7), ID number 6, which is link information on the input data string (DS_I), is recorded (step-S77).

Next, the MPU 10 executes Step-S78.
The adjustment process of the data level value is executed. That is, as shown in FIG. 16, the MPU 10 outputs the input data sequence (DS_I)
(ID number 6) and the parent concept data sequence (D
S_P) (ID number 7) is compared with the data level value.
More specifically, the MPU 10 stores the data level value area (Le) of the input data string stored in the variable X in the main storage device 14.
vel_X) is compared with the value of the data level value area (Level_Y) of the parent concept data string stored in the variable Y (step-S80). Note that the variables X and Y have the same structure as the variables X and Y described in the fourth embodiment. Then, the value of the data level value area (Level_Y) ≦ the data level value area (Level_Y)
For the value of (X), 1 is added to the value of the data level value area (Level_X).
Is stored in the data level value area (Level_Y) (step-S81). In the tenth embodiment, the value of the data level value area (Level_Y) is
el_Y), so the data level value area (Level_
Save “2” in Y).

After that, the contents of the variables X and Y are written in the data string of the corresponding ID number in the data structure (step-S82).

Next, the parent concept data string (DS_P) (ID
A data string is extracted based on the link information recorded in the parent data area No. 7). If the data level value of the extracted data string is equal to or less than the data level value adjusted in step S78, the data string is extracted. In the data level value area of
In step S78, a value obtained by adding 1 to the data level value adjusted is recorded. That is, when the data level value of the data string of ID number 7 changes from “1” to “2”, I
The data level value of the data string (ID number 12 in the tenth embodiment) in which the data identification name indicating the set including the data identification name recorded in the data string of D number 7 changes. Furthermore, the data sequence of the data sequence (in the tenth embodiment, the data sequence of the ID number 15 and the ID number 16) in which the data identification name indicating the set including the data identification name recorded in the data sequence of the ID number 12 is recorded. The level value also changes.
Therefore, the data level value readjustment process (step-S7)
9) is required.

In such a case, the data string is extracted based on the link information recorded in the parent data area in the data string in which the data identification name corresponding to the parent concept name is recorded, and the data string is extracted in the parent data area of the extracted data string. Based on the recorded link information, data strings are extracted until the parent data area of the extracted data string becomes empty. Then, the data level value of the extracted data sequence (DS-E) is
If the data level value of the data string (DS-F) extracted based on the link information recorded in the parent data area of E) is equal to or greater than 1, the data level value of the extracted data string (DS-E) is set to 1. The added value is recorded in the data level value area of the data string (DS-F) extracted based on the link information recorded in the parent data area.

For this purpose, steps S90 to S90 of the fourth embodiment described with reference to FIGS.
Step 102 may be executed. Thus, link information addition for adding new link information to the parent data area of the data string is completed.

(Embodiment 11) The eleventh embodiment is different from the tenth embodiment.
And relates to a link information adding method according to the fourth aspect of the present invention.

In the tenth embodiment which is the link information adding method according to the third aspect of the present invention, in step S75, the data stream is created based on the link information recorded in the parent data area of the parent concept data stream (DS_P). Was extracted. That is, this step is substantially the same as step-S1 of the first embodiment.
3 to Step-S16. On the other hand, Example 11
In the link information adding method, the data string is extracted based on the link information recorded in the element data area of the input data string (DS_I) in step S75.

That is, in the link information adding method according to the eleventh embodiment, in step S 74, the input data sequence (DS_P) is stored in the element data area of the parent concept data sequence (DS_P).
_I) is checked to determine whether link information exists.
Then, if the link information exists, the addition of the link information is stopped because the above rule is violated. This determination can also be made by determining whether or not link information on the parent concept data sequence (DS_P) is recorded in the parent data area of the input data sequence (DS_I).

If there is no link information on the input data string (DS_I) in the element data area of the parent concept data string (DS_P), the rule 'is checked. That is,
A data string is extracted based on the link information recorded in the element data area of the input data string (DS_I), and the element data of the extracted data string is extracted based on the link information recorded in the element data area of the extracted data string. Data strings are extracted until the area becomes empty. Specifically, the MPU 10
The element identification number area (Element) of the element data area (Element_X) of the data string stored in the variable X in the main storage device 14
_ID_X), the search pattern 3 or the search pattern 4 described above is executed. Since the data level value of the data string of ID No. 6 in which the data identification name “flammable” is recorded is “1”, in this search step,
Data strings are not extracted.

Then, one of the data strings extracted in this step is the data string (DS_DS) recording the parent concept name.
P) (in the eleventh embodiment, the ID number 7), the addition of the link information is stopped, and any of the data strings extracted in this step is replaced with the data string (DS_P ), The input data string (D
In the parent data area of (S_I) (ID number 6), an ID number 7 as link information relating to a data string (DS_P) in which a parent concept name is recorded is recorded (step-S77). In addition, the input data string (DS_P) is stored in the element data area in the data string (DS_P) (ID number 7) in which the parent concept name is recorded.
_I) is recorded as ID number 6, which is link information.

Except for these points, the other steps in the eleventh embodiment can be the same as those in the tenth embodiment, and thus detailed description is omitted.

Example 12 Example 12 relates to a link information adding method according to the fifth aspect of the present invention, and relates to a link information adding method for adding new link information to an element data area of a data string. The structure of the data string in the data structure of the twelfth embodiment and various data recorded in each data string were the same as in the first embodiment. The link information adding method according to the twelfth embodiment is basically similar to the second link information adding method according to the present invention described in the seventh embodiment. Therefore, the difference from the second link information adding method of the present invention described in the seventh embodiment is focused on, and again, FIGS.
Referring to the flowcharts of FIGS. 17 and 18,
A link information adding method according to the twelfth embodiment will be described.

As shown in FIG. 25, first, a data identification name is input from the input device 12 composed of, for example, a keyboard, and at the same time, an element concept name which is an element of a set represented by the data identification name is input ( Step-S140).
In the twelfth embodiment, in the data string of ID number 7 (input data string) in which the data identification name “plant” is recorded, the data identification name “flammable” is recorded as new link information.
Link information on the data sequence of D number 6 (element concept data sequence) is added. The MPU 10 reads the input character strings “plant” and “flammable” and stores them in a variable area set in the main storage device 14. And the MPU 10
It is checked whether or not the input data identification name matches the input element concept name (step-S141).
Since the above rule is violated, the addition of the link information is stopped.

If the input data identification name does not match the input element concept name, the input data sequence (DS_I) (ID
No. 7) is extracted, and an element concept data string (DS_E) (ID number 6), which is a data string in which a data identification name that matches the input element concept name is recorded (step S142, step S142). -S143).

Next, it is checked whether or not link information on the element concept data string (DS_E) exists in the element data area of the input data string (DS_I) (step-S14).
4). If there is, the addition of the link information is stopped because the above rule is violated. Such a determination can also be made by determining whether or not link information relating to the input data sequence (DS_I) is recorded in the parent data area of the element concept data sequence (DS_E).

If there is no link information on the element concept data string (DS_E) in the element data area of the input data string (DS_I), the rule 'is checked. That is, the data string is extracted based on the link information recorded in the parent data area of the input data string (DS_I), and the extracted data string is extracted based on the link information recorded in the parent data area of the extracted data string. Until the parent data area is empty
A data string is extracted (step-S145). This step is different from the seventh embodiment, and can be substantially the same as steps S13 to S16 in the first embodiment. In this search step, data strings of ID number 12, ID number 15, and ID number 16 are extracted.

If any one of the data strings extracted in this search step matches the data string (DS_E) (ID No. 6) in which the element concept data identification name is recorded, it violates the rule ', so that the link information Is stopped (step-S146).

If none of the data strings extracted in this search step matches the data string (DS_E) (ID number 6) recording the element concept data identification name, the input data string (DS_I) (ID number 7) ) Is recorded in the element data area of ID number 6 which is link information relating to the element concept data sequence (DS_E) (step-S147).
In addition, the element concept data string (DS_E) (ID number 6)
Is recorded in the parent data area of ID number 7 as link information relating to the input data string (DS_I) (step-S147). This step can be substantially the same as Step-S147 of the seventh embodiment.

Next, the MPU 10 executes Step-S14
8 is performed. That is, FIG.
As shown in FIG. 3, the MPU 10 receives the input data sequence (DS_
I) The data level value of (ID number 7) is compared with the data level value of the element concept data sequence (DS_E) (ID number 6). More specifically, the MPU 10
Then, the value of the data level value area (Level_X) of the input data string stored in the variable X is compared with the value of the data level value area (Level_Y) of the element concept data string stored in the variable Y (step- S150). Note that the variables X and Y have the same structure as the variables X and Y described in the seventh embodiment. If the value of the data level value area (Level_X) ≦ the value of the data level value area (Level_Y), the data level value area (Level_X)
_Y) plus 1 to the data level value area (Level_
X) (step-S151). In the twelfth embodiment, since the value of the data level value area (Level_X) is equal to the value of the data level value area (Level_Y), “2” is stored in the data level value area (Level_X).

After that, the contents of the variables X and Y are written in the data string of the corresponding ID number in the data structure (step-S152).

Next, the data string (DS_
If a data string is extracted based on the link information recorded in the parent data area of I), and the data level value of the extracted data string is equal to or less than the data level value stored in the data level value area (Level_X), In the data level value area of the extracted data string, a value obtained by adding 1 to the data level value stored in the data level value area (Level_X) is recorded.

That is, a data string is extracted based on the link information recorded in the parent data area of the input data string (DS_I) (ID number 7), and the data level value of the extracted data string is determined in step S151. If it is less than or equal to the adjusted data level value,
In step S151, a value obtained by adding 1 to the data level value adjusted is recorded. That is, by changing the data level value of the data string of ID number 7 from “1” to “2”,
The data level value of the data string (ID number 12 in the twelfth embodiment) in which the data identification name indicating the set including the data identification name recorded in the data string of ID number 7 changes. Further, the data sequence of the data sequence (in the twelfth embodiment, the data sequence of the ID number 15 and the ID number 16) recording the data identification name indicating the set including the data identification name recorded in the data sequence of the ID number 12 The level value also changes. Therefore, a readjustment process of the data level value is required.

In such a case, the data string is extracted based on the link information recorded in the parent data area in the parent concept data string, and the extracted data string is extracted based on the link information recorded in the parent data area of the extracted data string. The data string is extracted until the parent data area of the data string becomes empty. Then, the data level value of the extracted data string (DS-G) is calculated based on the data string (DS-H) extracted based on the link information recorded in the parent data area of the data string (DS-G). If it is equal to or higher than the level value, the extracted data string (DS-
G) is obtained by adding 1 to the data level value of the data sequence (D) extracted based on the link information recorded in the parent data area.
(S−H) in the data level value area.

For this purpose, steps S90 to S90 of the fourth embodiment described with reference to FIGS.
Step 102 may be executed. In this case, in step S
Before the execution of 90, the contents of variable X are moved to variable Y. Thus, link information addition for adding new link information to the element data area of the data string is completed.

(Thirteenth Embodiment) A thirteenth embodiment is different from the twelfth embodiment.
And relates to a link information adding method according to the sixth aspect of the present invention.

In Embodiment 12 which is the link information adding method according to the fifth aspect of the present invention, Step-S145
In, the data string was extracted based on the link information recorded in the parent data area of the input data string (DS_I). That is, this step is substantially the same as step-S1 of the first embodiment.
3 to Step-S16. On the other hand, Example 13
In the link information adding method of (1), Step-S145
, A data string is extracted based on the link information recorded in the element data area of the element concept data string (DS_E).

That is, in the link information adding method of the thirteenth embodiment, in step S 145, the input data sequence (D) is stored in the parent data area of the element concept data sequence (DS_E).
If there is no link information about the element concept data string (DS_E) in the element data area of the input data string (DS_I), the element data of the element concept data string (DS_E) does not exist. The data string is extracted based on the link information recorded in the area, and the data string is extracted until the element data area of the extracted data string becomes empty based on the link information recorded in the element data area of the extracted data string. Is extracted. Specifically, the MPU 10 determines the search pattern 3 or the search pattern 3 described above based on the data of the element identification number area (Element_ID_Y) of the element data area (Element_Y) of the data string stored in the variable Y in the main storage device 14. Execute search pattern 4. Since the data level value of the data string of ID number 6 in which the data identification name “flammable” is recorded is “1”, the data string is not extracted in this search step.

[0313] One of the data strings extracted in this step is the input data string (DS_I) (Embodiment 1).
In step 3, when the ID number matches the ID number 7) (step-S146), the addition of the link information is stopped, and none of the data strings extracted in this step match the input data string (DS_I). , The input data string (DS_
I) In the element data area of (ID number 7), ID number 6 which is link information on the element concept data sequence (DS_E)
Is recorded (step-S147). At the same time, an ID number 7 which is link information on the input data string (DS_I) is recorded in the parent data area in the element concept data string (DS_E) (ID number 6).

Except for these points, the other steps in the thirteenth embodiment can be the same as those in the twelfth embodiment, and thus detailed description is omitted.

(Embodiment 14) Embodiment 14 relates to a data search method according to the first and fourth aspects of the present invention, and more specifically, the first and fourth aspects of the present invention in the field of physician diagnosis. About. Table 2 shows the data structure in the fourteenth embodiment.
3 is shown.

[0316]

[Table 23]

[0317] It is very useful for diagnosis if a physician can know a possible disease from symptoms and test data shown by a patient. If the physician wants to know what the concept of leukocytosis, cough, and fever are the elementary and / or indirect elemental concepts at the same time, the parent concept and / or the concept of leukocytosis, fever, and cough are simultaneously present. It suffices if a data string having a data identifier that can be an indirect parent concept can be extracted from the data structure. That is, "white blood cell increase", "fever", and "cough" are input as data identification names to be searched. Table 24 below shows the result of the data search performed by the scrap-type data search method. Refer to the third embodiment for a specific data search process.

[0318]

[Table 24] Extracted data Number of data strings of input data strings that were hit Data identifiers of bacterial pneumonia 3 Viral pneumonia 2 Respiratory symptoms 1

As a result of the data search, it is found that a concept having all of “leukocytosis”, “fever” and “cough” as the element concept and / or the indirect element concept is “bacterial pneumonia”. Here, “cough” is not included in the element concept of “bacterial pneumonia”. However, it can be seen that "bacterial pneumonia" has "cough" as an indirect element concept because "respiratory symptoms" has "cough" as an element concept.

The same search result can be obtained by performing a data search by expressing all diseases with a data string of data level value “2”. However, in that case, the element set becomes a list of the symptoms and abnormalities of the test recorded in the first level data string, and it becomes difficult to grasp the characteristics of the disease. It is more reasonable to express the disease as much as possible with a medical concept that can express the features, and to express the used medical concept with yet another medical concept.

In the diagnosis, it is necessary to consider the case where the patient does not show typical symptoms. In such a case, for example, it is necessary to expand the search range to a disease having at least two of “white blood cell increase”, “fever”, and “cough” as an element concept and / or an indirect element concept. Looking at the data identifiers in which the number of input data strings hit in the search results in Table 24 is “2”, it can be seen that such diseases include “viral pneumonia” and “bacterial pneumonia”.

When searching for a disease having a respiratory symptom, the doctor needs only a disease having one of “cough” and “sputum” as an element concept and / or an indirect element concept. There is. For example, "Viral pneumonia"
The concept includes only the concept of "cough" out of "cough" and "sputum". Essentially, the set of “viral pneumonia” should be “fever, leukopenia, cough, sputum” or “fever, leukopenia, respiratory symptoms”. However, such coherency is not always maintained at the stage of forming the data structure.

By applying the data search method according to the fourth aspect of the present invention described in the ninth embodiment, for example, at least one element concept and / or indirect element concept of “respiratory symptoms” And / or a concept having the concept as an indirect element concept can be obtained. That is, as a result of the data search, “viral pneumonia” and “bacterial pneumonia” can be obtained. However, the display of "respiratory symptoms" and the first level data string is excluded.

Further, there is a problem of synonyms. For example, various expressions such as “fever”, “high fever”, and “slight fever” are used, and these three terms are properly used depending on the disease. Here, consider a case of searching for a febrile disease. In the conventional search method, it is necessary to perform a search by connecting "heat", "high heat" and "low heat" with "OR". Anyway,
In the data search method of the present invention, heat generation {high heat, low heat}, which is a data sequence corresponding to a concept including similar concepts, is created in advance. When the data search method according to the fourth aspect of the present invention is executed, “heat generation”, “high heat”,
All febrile diseases having “low fever” as an element concept and / or an indirect element concept can be obtained. in this way,
By creating a data string defined as a concept including synonyms, omission of keyword specification at the time of data search and complexity of specifying items are eliminated.

(Embodiment 15) Embodiment 15 relates to a data search method according to the second aspect of the present invention. In Example 15, the following chemical reaction is considered. MgO + H ₂ → Mg + H ₂ O C + O ₂ → CO ₂ CO ₂ + H ₂ O → H ₂ CO ₃

The data structure in Embodiment 15 is shown in Table 25 below.

[0327]

[Table 25]

In Table 25, a compound which is a data identifier recorded in a certain data string having a certain data level value corresponds to a data string corresponding to the link information recorded in the element data area of the certain data string. It is generated by the reaction of the compound or element which is the data identification name recorded in the.

In the fifteenth embodiment, “H ₂ CO ₃ ” is input as data to be searched. Then, after the same process as that of the fifth embodiment, if only the data string having the data level value “1” is displayed as the data search result,
"MgO", "H _2", "C", can be obtained "O _2". This result indicates that H ₂ CO ₃ is MgO, H ₂ , C, O ₂
Is generated from

Although the present invention has been described based on the preferred embodiments, the present invention is not limited to these embodiments. The structure of the data string in the data structure is an example, and can be changed as appropriate. The structures of the search data array and the search element data array are also examples, and can be changed as appropriate. Furthermore, the method of extracting a data string is also an example, and any method may be used.

For example, a structure as shown in FIG. 33 can be adopted. The data structure having the structure shown in FIG. 33 includes a conceptual structure, a parent set structure, and an element set structure.
Each data string in the conceptual structure has an identification number area (I
D), name area (Name), level data area (Level), parent data area (Parent), and element data area (Elemen
t). An identification number for identifying a data string is recorded in the identification number area (ID), and this number is a number unique to each data string. Name space (Na
In me), the concept name of the corresponding level (first level concept name, second level concept name, third level concept name, etc.) is recorded.
The level (1, 2, 3, etc.) of the data string is recorded in the level data area (Level_Value).

The parent data area (Parent) stores the number of parent concept link information.
_Count), the parent aggregate structure link information first area (Parent_LinkT) storing the ID number of the head of the related parent aggregate structure
op), parent aggregate structure link information second area (Parent_LinkEnd) storing the last ID number of the associated parent aggregate structure
It is composed of If there is no related parent aggregate structure, the parent aggregate structure link information first area (Parent_LinkT)
op) and parent aggregate structure link information second area (Parent_Lin)
“void” is written in kEnd).

The element data area (Element) includes an element concept link information count area (Element_Count) for storing the number of element concept link information, and an element set structure link information for storing a head ID number of a related element set structure. A first area (Element_LinkTop) and an element set structure link information second area (Element_LinkEnd) for storing the last ID number of the related element set structure. If there is no related element set structure, “void” is written in the element set structure link information first area (Parent_LinkTop) and the element set structure link information second area (Parent_LinkEnd).

The parent set structure includes an ID number area (ID) for storing an ID number, a next ID number area (Next_ID) for storing the ID number of the next parent set structure, and a parent set structure. A parent ID number area (Before_ID) for storing the ID number of the parent aggregate structure in front of the parent set structure, and a parent identification number area (Parent_ID). The next parent aggregate structure is used when the parent aggregate structure alone cannot store the parent concept link information. Parent identification number area (Pare
In nt_ID), an ID number as parent concept link information is recorded.

An element set structure includes an ID number area (ID) for storing an ID number, a next ID number area (Next_ID) for storing an ID number of the next element set structure, and an element set structure. A preceding ID number area (Before_ID) and an element identification number area (Element_ID) for storing the ID number of the element set structure located before. The next element set structure is used when the element concept link information alone cannot store the element concept link information. In the element identification number area (Element_ID), an ID number as element concept link information is recorded.

In a data structure having such a structure, by inputting a data identifier, a data string (input data string) of a conceptual structure in which the data identifier is recorded is read out. The parent aggregate structure data can be reached based on the ID number recorded in the parent aggregate structure link information first area (Parent_LinkTop) of the column. Then, by reading the contents of the data of the parent aggregate structure, link information relating to the parent concept can be obtained. The link information is in the parent identification number area (Parent_I
If it cannot be accommodated by only D), the next ID number area (Next_I) for storing the ID number of the next parent aggregate structure
Based on the ID number recorded in D), the data of the next parent aggregate structure can be read. The reading of the link information of the element concept in the element set structure is the same.

[0337]

According to the data search method or the link information addition method of the present invention, it is possible to completely prevent the possibility of falling into an infinite loop in the search and completely eliminate the cycle definition. Therefore, when adding the link information in the data structure, the user can add the link information without paying any attention to the occurrence of the cyclic definition. When searching for data,
An excellent effect that not only the extraction of the parent concept and the element concept but also the extraction of the indirect parent concept and the indirect element concept can be performed.

According to the present invention, it is possible to construct a data structure for efficiently and easily expressing knowledge in a hierarchical structure. In such a data structure, all knowledge is modeled by a set having a name called a concept,
The user can freely construct and reconstruct a data structure having a hierarchical structure. Direct or indirect cyclic definitions that occur when adding one concept to another as an element concept can be automatically eliminated.

[Brief description of the drawings]

FIG. 1 is a diagram schematically showing a search pattern 1 in a data search method according to a first embodiment of the present invention.

FIG. 2 is a diagram schematically showing another search pattern 2 in the data search method according to the first embodiment of the present invention.

FIG. 3 is a diagram schematically showing a search pattern 3 in a data search method according to a second embodiment of the present invention.

FIG. 4 is a diagram schematically showing a search pattern 4 in the data search method according to the second embodiment of the present invention.

FIG. 5 is a conceptual diagram showing a structure of a data string of each level and a structure of a first-level data string in the first embodiment.

FIG. 6 is a conceptual diagram showing a structure of second and third level data strings in the first embodiment.

FIG. 7 is a conceptual diagram showing a structure of second and third level data strings in the first embodiment.

FIG. 8 is a flowchart illustrating a data search method according to the first embodiment.

FIG. 9 is a search process 1 in the data search method according to the first embodiment.
5 is a flowchart for explaining FIG.

FIG. 10 is a flowchart illustrating a data search method according to a second embodiment.

FIG. 11 is a flowchart illustrating search processing 2 in the data search method according to the second embodiment.

FIG. 12 is a diagram illustrating an execution procedure of search processing 2 in the data search method according to the second embodiment.

FIG. 13 is a flowchart illustrating a data search method according to a third embodiment.

FIG. 14 is a flowchart illustrating the data search method according to the third embodiment, following FIG. 14;

FIG. 15 is a flowchart illustrating a link information adding method according to a fourth embodiment.

FIG. 16 is a flowchart illustrating adjustment of a data level value in the link information adding method according to the fourth embodiment.

FIG. 17 is a flowchart for explaining a link information adding method according to the fourth embodiment, following FIG. 15;

FIG. 18 is a flowchart for explaining a link information adding method according to the fourth embodiment, following FIG. 17;

FIG. 19 shows a structure of a data string of each level according to the fifth embodiment;
FIG. 4 is a conceptual diagram showing a structure of a first level data string.

FIG. 20 is a conceptual diagram showing a structure of second and third level data strings in the fifth embodiment.

FIG. 21 is a conceptual diagram showing a structure of second and third level data strings in the fifth embodiment.

FIG. 22 is a flowchart illustrating a data search method according to a fifth embodiment.

FIG. 23 is a flowchart illustrating search processing 3 in the data search method according to the fifth embodiment.

FIG. 24 is a flowchart illustrating a data search method according to a sixth embodiment.

FIG. 25 is a flowchart illustrating a link information adding method according to a seventh embodiment.

FIG. 26 is a flowchart illustrating adjustment of a data level value in a link information adding method according to a seventh embodiment.

FIG. 27 is a conceptual diagram showing a structure of a data string of each level according to the eighth embodiment.

FIG. 28 is a conceptual diagram showing a structure of a first level data string in an eighth embodiment.

FIG. 29 is a conceptual diagram showing a structure of a second level data string in the eighth embodiment.

FIG. 30 is a conceptual diagram showing a structure of a third level data string in the eighth embodiment.

FIG. 31 is a conceptual diagram showing a structure of a second level data string in the eighth embodiment.

FIG. 32 is a conceptual diagram showing the structure of a third level data string in the eighth embodiment.

FIG. 33 is a conceptual diagram showing another example of the structure of a data string of each level according to the present invention.

FIG. 34 is a conceptual diagram showing an apparatus suitable for executing the data search method of the present invention.

[Explanation of symbols]

 Reference Signs List 10 MPU 12 Input device 14 Main storage device 16 Display device 18 External storage device

──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-6-52232 (JP, A) JP-A-3-58130 (JP, A) JP-B-7-7350 (JP, B2) Nick Roussoopoulos, Hyun Soon Kim, "R OOST: A Relational Object Oriented System", Lect Notes Computer Sci Vol. 367 pp. 404-420, (Jicst received on July 11, 1989) CN: 89A0487877 Yoshifumi Masunaga, "Introduction to Relational Databases", pp. 28-30, Science Inc. (First edition issued on January 25, 1991) (58) Fields investigated (Int. Cl. ⁷ , DB name) G06F 17/30 G06F 12/00 512

Claims (6)

(57) [Claims] 1. Each data string (DS_1) includes: (a) a data identifier name area in which a data identifier is recorded; (b) a data level value area in which a data level value is recorded; A parent identification number area for recording link information on a data string (DS_2) recording a data identification name indicating a set including the data identification name is stored in the data string (DS_2).
Only a few or more number, parent data area having, consist, data level values of (A) data strings (DS_2) is greater than the data level values of the data sequence (DS_1), (B) data strings The data string (DS_1) is stored in any of the parent identification number areas constituting the parent data area of (DS_1).
All the link information recorded in the parent identification number area forming the parent data area of the data string (DS_1) does not include the link information relating to itself, and (C) forms the parent data area of the data string (DS_1). Starting from the link information recorded in an arbitrary parent identification number area, the parent identification number area constituting the parent data area of the data string is kept until the parent identification number area constituting the parent data area of the data string becomes empty. When the data sequence is sequentially traced based on the recorded link information, the link information relating to the data sequence (DS_1) is not included in the parent identification number region constituting the parent data region of any traced data sequence. Link information adding method for adding new link information to a parent identification number area constituting a parent data area of a data string in a data string aggregate composed of data strings (A) inputting a data identifier and inputting a parent concept name indicating a set including the data identifier; and (b) a parent concept having the input data identifier. (C) If the input data identification name does not match the input parent concept name, the data string (DS_I) recording the input data identification name is deleted. (D) extracting a data string (DS_P) that records a data identifier matching the input parent concept name; and (e) extracting the data string (DS) extracted in step (c). DS_I)
If link information relating to the data string (DS_P) extracted in step (d) exists in the parent identification number area constituting the parent data area of step (c), the addition of link information is stopped, and (f) step (c) (DS_I)
If there is no link information on the data string (DS_P) extracted in the step (d) in the parent identification number area that constitutes the parent data area of the data string (DS_P) extracted in the step (d), The data string is extracted based on the link information recorded in the parent identification number area constituting the parent data area, and extracted based on the link information recorded in the parent identification number area constituting the parent data area of the extracted data string. (G) extracting a data string until the parent identification number area constituting the parent data area of the extracted data string becomes empty; and (g) any one of the data strings extracted in step (f) comprises: If it matches the data string (DS_I) extracted in c), the addition of the link information is stopped, and (h) any of the data strings extracted in step (f) is changed in step (c). Extracted data string (DS If that does not match the I), if the lack of the number of parent identification number area
If there is, create a new parent identification number area,
A step of recording link information on the data string (DS_P) extracted in the step (d) in a parent identification number area constituting a parent data area of the data string (DS_I) extracted in the step (c); (I) Data string (DS_P) extracted in step (d)
Is less than or equal to the data level value of the data string (DS_I) extracted in the step (c), the data extracted in the step (c) is stored in the data level value area of the data string (DS_P). Recording a value obtained by adding 1 to the data level value of the column (DS_I). 2. Each data string (DS_1) includes: (a) a data identifier name area in which a data identifier is recorded; (b) a data level value area in which a data level value is recorded; An element identification number area for recording link information on a data string (DS_0) recording a data identification name which is an element of a set represented by the data identification name is stored in the data string.
( A) The data level value of the data string (DS_0) is smaller than the data level value of the data string (DS_1). ) The data string (DS_1) is placed in any of the element identification number areas constituting the element data area of the data string (DS_1).
1) All link information recorded in the element identification number area constituting the element data area of the data string (DS_1) does not include link information relating to itself, and (C) the element data area of the data string (DS_1) Starting from the link information recorded in an arbitrary element identification number area constituting an element identification number, an element identification number constituting an element data area of a data string until the element identification number area constituting an element data area of the data string becomes empty When the data sequence is sequentially traced based on the link information recorded in the area, the data sequence (DS_1)
The link information related to the element data area of the data string in the data string aggregate composed of a plurality of data strings, which is not included in the element identification number area constituting the element data area of any of the traced data strings A link information adding method for adding new link information to a constituent element identification number area, comprising the steps of: (a) inputting a data identification name, and simultaneously inputting an element concept name which is an element of a set represented by the data identification name; (B) if the input data identification name matches the input element concept name, stop adding link information; and (c) the input data identification name is the input element concept name. If not, a data string (DS_I) recording the input data identification name is extracted, and (d) a data string recording the data identification name that matches the input element concept name A step of extracting the DS_E), (e) step (c) extracted by data string (DS_I)
If link information relating to the data string (DS_E) extracted in the step (d) exists in the element identification number area constituting the element data area of (c), the addition of the link information is stopped, and (f) step (c) (DS_I)
If there is no link information on the data string (DS_E) extracted in the step (d) in the element identification number area constituting the element data area of (a), the data string (DS_E) extracted in the step (d) The data string is extracted based on the link information recorded in the element identification number area constituting the element data area, and extracted based on the link information recorded in the element identification number area constituting the element data area of the extracted data string. (G) extracting a data string until the element identification number area constituting the element data area of the data string obtained becomes empty; and (g) any one of the data strings extracted in the step (f) comprises: If it matches the data string (DS_I) extracted in c), the addition of the link information is stopped, and (h) any of the data strings extracted in step (f) is changed in step (c). Extracted If it does not match the data sequence (DS_I), insufficient if the number of element identification number field
If it is, create a new element identification number area
In, the element identification number field that constitutes the element data area of the process the extracted data sequence in (c) (DS_I), step of recording the link information about the process the extracted data sequence in (d) (DS_E) And (i) the data string (DS_E) extracted in step (d)
Is greater than or equal to the data level value of the data string (DS_I) extracted in step (c), the data level value recorded in the data string (DS_E) extracted in step (d) is 1 Recording the value to which is added in the data level value area of the data string (DS_I) extracted in the step (c). 3. Each data string (DS_1) includes: (a) a data identification name area in which a data identification name is recorded; (b) a data level value area in which a data level value is recorded; A parent identification number area for recording link information on a data string (DS_2) recording a data identification name indicating a set including a name is stored in the data string (DS_2).
The parent data area, which has or more than the number of
And, (d) the data identification name element identification number area for recording link information about an element in which data DN recorded data sequence of the set (DS_0) represented by,該De over data string
( A) The data level value of the data string (DS_2) is larger than the data level value of the data string (DS_1). ) The data string (DS_1) is stored in any of the parent identification number areas constituting the parent data area of the data string (DS_1).
All the link information recorded in the parent identification number area forming the parent data area of the data string (DS_1) does not include the link information relating to itself, and (C) forms the parent data area of the data string (DS_1). Starting from the link information recorded in an arbitrary parent identification number area, the parent identification number area constituting the parent data area of the data string is kept until the parent identification number area constituting the parent data area of the data string becomes empty. When the data sequence is sequentially traced based on the recorded link information, the link information on the data sequence (DS_1) is not included in the parent identification number region constituting the parent data region of any traced data sequence. ) The data level value of the data string (DS_0) is smaller than the data level value of the data string (DS_1), and (E) constitutes the element data area of the data string (DS_1). None of containing the identification number field, the data sequence (DS_
1) The link information related to itself is not included, and the link information recorded in the element identification number area constituting the element data area of the data string (DS_1) is all different. (F) The element data area of the data string (DS_1) Starting from the link information recorded in an arbitrary element identification number area constituting an element identification number, an element identification number constituting an element data area of a data string until the element identification number area constituting an element data area of the data string becomes empty When the data sequence is sequentially traced based on the link information recorded in the area, the data sequence (DS_1)
The link information related to the parent data area of the data string in the data string aggregate composed of a plurality of data strings, which is not included in the element identification number area constituting the element data area of any of the traced data strings A link information adding method for adding new link information to a constituent parent identification number area, comprising the steps of: (a) inputting a data identification name and inputting a parent concept name indicating a set including the data identification name; (B) if the input data identifier matches the input parent concept name, stop adding link information; and (c) match the input data identifier with the input parent concept name. Otherwise, the data string (DS_I) recording the input data identification name is extracted, and (d) the data string (DS_P) recording the data identification name that matches the input parent concept name is extracted. A step, (e) step (c) extracted by data string (DS_I)
If link information relating to the data string (DS_P) extracted in step (d) exists in the parent identification number area constituting the parent data area of step (c), the addition of link information is stopped, and (f) step (c) (DS_I)
If there is no link information relating to the data string extracted in step (d) in the parent identification number area constituting the parent data area of step (d), the data string (D
S_P), a data string is extracted based on the link information recorded in the parent identification number area constituting the parent data area, and the link information recorded in the parent identification number area constituting the parent data area of the extracted data string is extracted. Extracting a data string until the parent identification number area constituting the parent data area of the extracted data string becomes empty, and (g) one of the data strings extracted in step (f) , If it matches the data string (DS_I) extracted in step (c), the addition of link information is stopped, and (h) any of the data strings extracted in step (f) ) Does not match the data string (DS_I) extracted if the number of parent identification number areas is insufficient.
If there is, create a new parent identification number area,
In the parent identification number area constituting the parent data area of the data string (DS_I) extracted in the step (c), link information on the data string (DS_P) extracted in the step (d) is recorded. (I) the data string (DS_P) extracted in step (d)
If the number of element identification number areas is insufficient in the element identification number area that constitutes the element data area in
Is a step of recording link information relating to the data string (DS_I) extracted in step (c) after newly generating an element identification number area ; and (j) the data extracted in step (d). Column (DS_P)
Is less than or equal to the data level value of the data string (DS_I) extracted in the step (c), the data extracted in the step (c) is stored in the data level value area of the data string (DS_P). Recording a value obtained by adding 1 to the data level value of the column (DS_I). 4. The method according to claim 1, further comprising the step of : (e ′) extracting the data string (DS) extracted in the step (d) instead of the steps (e), (f), (g) and (h).
_P) in the element identification number area constituting the element data area, the data string (DS_
If link information relating to I) exists, the addition of the link information is stopped, and (f ′) the data string (DS) extracted in the step (d).
_P) in the element identification number area constituting the element data area, the data string (DS_
If there is no link information on I), step (c)
The data string is extracted based on the link information recorded in the element identification number area forming the element data area of the data string (DS_I) extracted in the above, and the element identification number forming the element data area of the extracted data string Extracting a data string based on the link information recorded in the area until an element identification number area constituting an element data area of the extracted data string becomes empty; and (g ′) step (f ′) One of the data strings extracted in step (d) is the data string (DS_P) extracted in step (d).
If it matches, the addition of the link information is stopped, and (h ′) any of the data strings extracted in step (f ′) is replaced with the data string (DS_P) extracted in step (d).
If they do not match, the number of parent identification number areas is insufficient.
If it has, create a new parent identification number area
Then, link information on the data string (DS_P) extracted in the step (d) is recorded in the parent identification number area constituting the parent data area of the data string (DS_I) extracted in the step (c). 4. The method according to claim 3 , further comprising the steps of: 5. Each data string (DS_1) includes: (a) a data identification name area in which a data identification name is recorded; (b) a data level value area in which a data level value is recorded; A parent identification number area for recording link information on a data string (DS_2) recording a data identification name indicating a set including a name is stored in the data string (DS_2).
The parent data area, which has or more than the number of
And (d) an element identification number area for recording link information relating to a data string (DS_0) recording a data identification name that is an element of a set represented by the data identification name,
( A) The data level value of the data string (DS_2) is larger than the data level value of the data string (DS_1). ) The data string (DS_1) is stored in any of the parent identification number areas constituting the parent data area of the data string (DS_1).
All the link information recorded in the parent identification number area forming the parent data area of the data string (DS_1) does not include the link information relating to itself, and (C) forms the parent data area of the data string (DS_1). Starting from the link information recorded in an arbitrary parent identification number area, the parent identification number area constituting the parent data area of the data string is kept until the parent identification number area constituting the parent data area of the data string becomes empty. When the data sequence is sequentially traced based on the recorded link information, the link information on the data sequence (DS_1) is not included in the parent identification number region constituting the parent data region of any traced data sequence. ) The data level value of the data string (DS_0) is smaller than the data level value of the data string (DS_1), and (E) constitutes the element data area of the data string (DS_1). None of containing the identification number field, the data sequence (DS_
1) The link information related to itself is not included, and the link information recorded in the element identification number area constituting the element data area of the data string (DS_1) is all different. (F) The element data area of the data string (DS_1) Starting from the link information recorded in an arbitrary element identification number area constituting an element identification number, an element identification number constituting an element data area of a data string until the element identification number area constituting an element data area of the data string becomes empty When the data sequence is sequentially traced based on the link information recorded in the area, the data sequence (DS_1)
The link information related to the element data area of the data string in the data string aggregate composed of a plurality of data strings, which is not included in the element identification number area constituting the element data area of any of the traced data strings A link information adding method for adding new link information to a constituent element identification number area, comprising the steps of: (a) inputting a data identification name, and simultaneously inputting an element concept name which is an element of a set represented by the data identification name; (B) if the input data identification name matches the input element concept name, stop adding link information; and (c) the input data identification name is the input element concept name. If not, a data string (DS_I) recording the input data identification name is extracted, and (d) a data string recording the data identification name that matches the input element concept name A step of extracting the DS_E), (e) step (c) extracted by data string (DS_I)
If link information relating to the data string (DS_E) extracted in the step (d) exists in the element identification number area constituting the element data area of (c), the addition of the link information is stopped, and (f) step (c) (DS_I)
If there is no link information on the data string (DS_E) extracted in the step (d) in the element identification number area constituting the element data area of (a), the data string (DS_I) extracted in the step (c) The data string is extracted based on the link information recorded in the parent identification number area constituting the parent data area, and extracted based on the link information recorded in the parent identification number area constituting the parent data area of the extracted data string. (G) extracting a data string until the parent identification number area constituting the parent data area of the extracted data string becomes empty; and (g) any one of the data strings extracted in step (f) comprises: If the data string (DS_E) extracted in d) matches, the addition of link information is stopped, and (h) any of the data strings extracted in step (f) is replaced in step (d). The extracted data column ( If S_E) does not match, insufficient if the number of element identification number field
If it is, create a new element identification number area
Then, link information on the data string (DS_E) extracted in the step (d) is recorded in the element identification number area constituting the element data area of the data string (DS_I) extracted in the step (c), In addition, (i) the data string (DS_E) extracted in step (d)
The parent ID number area constituting the parent data area in, also
If the number of parent identification number areas is insufficient, a new
Generating a parent identification number area and recording link information on the data string (DS_I) extracted in step (c); and (j) data string (DS_E) extracted in step (d). )
Is greater than or equal to the data level value of the data string (DS_I) extracted in step (c), the data level value recorded in the data string (DS_E) extracted in step (d) is 1 Recording the value to which is added in the data level value area of the data string (DS_I) extracted in the step (c). 6. The method according to claim 6, wherein in place of the steps (e), (f), (g) and (h), (e ′) the data string (DS) extracted in the step (d)
_E), if link information relating to the data string (DS_I) extracted in the step (c) exists in the parent identification number area constituting the parent data area of (E), stop adding the link information; (f ′) The data string (DS) extracted in the step (d)
_E), if there is no link information related to the data string (DS_I) extracted in the step (c) in the parent identification number area constituting the parent data area of the data string (DS) extracted in the step (d) DS_E), a data string is extracted based on the link information recorded in the element identification number area constituting the element data area, and the link information recorded in the element identification number area constituting the element data area of the extracted data string is extracted. (G ′) extracting the data string until the element identification number area constituting the element data area of the extracted data string becomes empty, or (g ′) the data string extracted in the step (f ′) Is the data string (DS_I) extracted in step (c)
If it matches, the addition of the link information is stopped, and (h ′) any of the data strings extracted in step (f ′) is replaced with the data string (DS_I) extracted in step (c).
If they do not match, the number of element identification number areas is insufficient
If it is, a new element identification number area is generated
Above, the data string (DS_I) extracted in step (c)
To the component data area element identification number region constituting the A link according to claim 5, characterized in that it comprises a step, to record link information about the extracted data sequence (DS_E) in said step (d) How to add information.