KR101693429B1 - System for identifying human relationships around users and coaching based on identified human relationships - Google Patents

Abstract

A coaching information providing system is disclosed through identification of an interpersonal relationship type. The method of providing coaching information through identification of an interpersonal relationship type of an interpersonal coaching system according to an embodiment includes confirming relationship data relating to a relationship between a first object and a second object having an interpersonal relationship with the first object And generating an item-by-item record of a data field table for defining an interpersonal relationship type from the relationship data, and determining an interpersonal type relationship between the first object and the second object based on the record do.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to coaching information providing systems,

The present invention relates to an interpersonal coaching information providing system, and more particularly, to a coaching information providing system through interpersonal relationship type grasp.

In the prior art, there are various recommendation methods such as content-based recommendation, collaborative filtering recommendation, and the like, but the recommendation for the product for the user himself remains.

For example, Korean Patent Laid-Open Publication No. 2012-0108536 discloses a service for providing anniversary notification service using a smart phone and providing product recommendation information including items of interest and price information of the user.

Korean Unexamined Patent Application Publication No. 2009-0078561 discloses a wire / wireless hybrid network system, which includes an application server for storing shopping information and product information, a personal server for collecting personal events related to the acquaintance of a subscriber, An intelligent content recommendation and delivery system for judging a preferred product, receiving information of the preferred product from the application server and determining the preferred product of the acquaintance, receiving information of the preferred product from the application server, and transmitting the intelligent content recommendation and delivery system, And an IP multimedia subsystem that receives information on the preferred products of the event and the acquaintance and provides the information to the mobile communication network based on the Internet Protocol (IP).

However, Korean Patent Laid-Open No. 2009-0078561 merely performs a product recommendation using the event information on the acquaintance of the subscriber, but does not take into account the relationship between the acquaintance and the various relationship types.

Korean Patent Publication No. 2006-0073333, Korean Patent No. 866,080, and Korean Patent Publication No. 2008-0092645 disclose a model for checking a communication period with an acquaintance using a terminal of a user, Only models that perform simple notification depend on the information.

Korean Unexamined Patent Publication No. 2006-0073333 (entitled " Mobile communication terminal for notifying the user of incoming / outgoing history data for human network management and method thereof) Korean Patent No. 866,080 (entitled " Intimacy Management < RTI ID = 0.0 > Korean Patent Publication No. 2009-0078561 (entitled " Intelligent Anniversary Notice & Product Recommendation System & Method, and Wired / Wireless Composite Network System Using It) Korean Patent Publication No. 2008-0092645 (entitled " Method and Apparatus for Providing Human Relation Information through Analysis of Log Data of Personal Communication Terminal) Korean Patent Laid-Open Publication No. 2012-0108536 (Name of invention: anniversary notification using smartphone and method of providing product information and apparatus thereof)

It is an object of the present invention to provide a system capable of converging a cooperative service with a relational network of service users or objects.

In addition, embodiments of the present invention provide a method and system that can infer cooperative types of a service user or an object relative to nearby persons and provide coaching based on a cooperative relationship type.

For example, the embodiment of the present invention can be applied to a user who can cooperate with a user in a range of a user's personal relationship by utilizing a relationship between the user and a user, Method and system.

In addition, the embodiment of the present invention can grasp the object to be applied to the behavior, grasp the execution timing of the behavior, grasp the recommendation acceptance level and scope of the behavior practitioner, grasp the coach information providing path and method preferred by the behavior practitioner, And to provide a method and system that can provide.

 According to an embodiment of the present invention, there is provided a method for providing coaching information through identification of an interpersonal relationship type of an interpersonal coaching system, The method comprising the steps of: confirming relationship data relating to a relationship between a first object and a second object having an interpersonal relationship; generating an item-by-item record of a data field table for defining an interpersonal relationship type from the relationship data; Determining a type of interpersonal relationship between the first object and the second object based on the type of interpersonal relationship between the first object and the second object.

According to an embodiment of the present invention, an apparatus for providing coaching information through identification of an interpersonal relationship type includes an interpersonal relationship type database in which an interpersonal relationship type is defined, a data field table for defining the interpersonal relationship type, A record generating unit for generating a record for each item of the data field table from the relational data, and a record generating unit for generating a record for each item of the data field table based on the record, 1 < / RTI > object and the second object.

According to the present invention, a system model capable of converging a cooperative service with a relational network of a service user or an object can be provided.

Further, according to the embodiment of the present invention, it is possible to provide a method and a system that can infer a relationship type of a service user or an object with a surrounding relative and provide coaching based on a relationship type.

Also, it is possible to provide a method and system for coaching a behavior of a user who is in a range of a user's personal relationship by utilizing a relationship between a user and a user, beyond the limitation of the product recommendation of the user himself .

Also, it is possible to provide a method and system for coaching a gift of a user who is preferred by a person within a range of an interpersonal relationship of the user by utilizing the relationship between the user and the user.

In addition, it is possible to identify the object to which the behavior is applied, to grasp the execution timing of the behavior, to grasp the level and range of recommendation acceptance of the behavior practitioner, to identify the path and method of coaching information preferred by the behavior practitioner, System can be provided.

For example, the terminal user can recognize the presence of a person who needs a gift among the surrounding people, and recommend the type of gift and the point of time when the gift recipient desires the desired gift, thereby contributing to the improvement of the interpersonal relationship of the service user.

1 is a diagram for explaining a configuration of a coaching system considering a relationship type according to an embodiment of the present invention.
2 is a diagram illustrating a configuration of a coaching service server according to an embodiment of the present invention.
3 is a flow chart illustrating a coaching method that takes into account the relationship type according to an embodiment of the present invention.
4 is a flowchart illustrating a coaching method considering a relationship type according to another embodiment of the present invention.
5 is a flowchart illustrating a method of determining a coaching method according to an embodiment of the present invention.
6A is a diagram illustrating a configuration of an apparatus for providing coaching information through identification of an interpersonal relationship type according to an embodiment of the present invention.
FIG. 6B shows an example of a configuration of an apparatus for calculating and determining an index according to an embodiment of the present invention.
FIG. 7A is a flowchart illustrating a method of providing coaching information through identification of an interpersonal relationship type in an interpersonal coaching system according to an exemplary embodiment of the present invention.
FIG. 7B shows an example of a procedure for selecting relationship data.
FIG. 7C shows a procedure for grasping a coaching level according to an embodiment.
8 and 9 are flowcharts for explaining an interpersonal relationship type determination method according to an embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

First, the concept of terms used throughout this specification will be described.

In the present specification, the term " relationship type "is not limited to a relationship between a human and a human, and in the situation where there are two objects under interaction, . ≪ / RTI >

For example, the relationship type may be an interpersonal relationship type.

In addition, the relationship type may be applied between devices and devices, for example, in a cloud computing environment or object Internet environment, where the objects may refer to individual devices.

For example, when it is difficult to grasp the needs, situations, and tendencies of the partner device in an environment where data acquisition and analysis are not easy to improve the relationship between the devices because the coaching target device and the coaching application device are not in the same control Embodiments of the invention may be applied.

For example, it is possible to collect data transmitted from a first device to a second device, to infer the relationship type of the first device and the second device, and to provide coaching information to the first device in consideration of the inferred relationship type have.

For example, the server may deduce from the data that the first device is the control subject and the second device is the controlled relationship type, and "coaching information that should be sent to the second device to shut down after a certain time" To the first device. At this time, the first device may or may not perform coaching according to predetermined conditions in consideration of a communication environment, a power condition, and surrounding environment information.

Hereinafter, for convenience of explanation, the relationship and the object will mainly be described with respect to interpersonal relationship and a user (for example, a service subscriber).

Coaching can be of various kinds. The interpersonal relationship may include providing gift coaching or interpersonal skills coaching information to improve interpersonal relationships.

Coaching can also include behavior coaching on behaviors of the object being coached, delivering a specific command or situation to the device, and psychological coaching. For example, behavioral coaching may be to present a gift by recommending a gift, and psychological coaching may be influencing the psychology of an object being coached through advice or counseling in interpersonal relationships or stages.

Also, the recommendation of the commodity, the action to be taken, and the like can be included in the coaching information. Furthermore, information on the accuracy and reliability of the provided coaching information may also be included.

Thus, in the prior art, rather than simply managing the intimacy of interpersonal relationships, embodiments of the present invention include inferring relationship types and providing coaching in consideration of relationship types.

If coaching information is provided to a robot or device with artificial intelligence, the coaching information may include instructions for communicating situations that may be a reference to the behavior of the robot or device, have.

Hereinafter, for the convenience of explanation, coaching can mainly be described as gift coaching related to recommendation of gifts, but interpersonal skills coaching may also be included.

1 is a diagram for explaining a configuration of a coaching system considering a relationship type according to an embodiment of the present invention.

Referring to FIG. 1, the coaching system includes subscriber terminals 110 and 120 and a coaching service server 140 that provides coaching services.

The subscriber terminals 110 and 120 and the coaching service server 140 may be connected through the communication network 130. [ At this time, the communication network 130 may include a mobile communication network such as a cloud computing environment, a wireless Internet, a local area network, and an LTE network.

2 is a diagram illustrating a configuration of a coaching service server according to an embodiment of the present invention.

Referring to FIG. 2, the coaching service server 200 includes a relationship data obtaining unit 220, a relationship type inference unit 230, a coaching providing method determining unit 240, and a coaching information providing unit 250. The coaching service server 200 may further include a communication unit 210 for collecting data through the communication network and for transmitting coaching information to a user terminal or the like and a database 260 for storing collected data and generated data have.

The relationship data acquiring unit 220 acquires relationship data related to the formation of a relationship between a first object to be coached and a second object related to the first object.

The relationship type inference unit 230 deduces the type of relationship between the first object and the second object based on the relationship data.

The coaching providing method determining unit 240 determines a coaching providing method considering the inferred relation type.

At this time, the coaching providing method determining unit 240 may include a first procedure for inferring a relationship forming step between the first object and the second object using the relation data, A third procedure for determining a coaching content, and a fourth procedure for grasping an acceptable coaching level of the first object.

The coaching information providing unit 250 provides coaching information to the first object according to a coaching providing method.

Meanwhile, when the coaching service server 200 determines that coaching is necessary to improve the interpersonal relationship between a specific person and the coaching target person among persons belonging to the interpersonal network of the coaching subject even though the first object as the coaching object is not recognized, .

For example, when the predetermined relationship type of the first object and the second object is derived as "partner company " and the promotion related keyword of the second object is extracted from the collected relationship data, And " date and intimacy "to the first object. At this time, for example, coaching may be a recommendation of a gift, or it may be an information provision that a celebration is needed. At this time, the promotion related keyword of the second object may be extracted from the promotion message or the text message information described in the SNS of the second object.

Therefore, the coaching service server 200 can identify the coaching subject based on the relationship data, and can identify the application target of coaching and the coaching application time corresponding to the coaching subject.

At this time, the coaching service server 200 grasps the level, range, and method of coaching that the coaching object can accommodate from the relationship data, and determines the coaching suitable for the appropriate time based on the level, And so on.

The relationship data obtaining unit 220, the relationship type inference unit 230, the coaching providing method determining unit 240, and the coaching information providing unit 250 shown in FIG. 2 may be installed not only in the server but also in the user terminal.

Hereinafter, embodiments of a coaching method considering a relationship type that can be performed by the coaching service server 200 will be described with reference to FIGS. 3 to 5. FIG.

3 is a flow chart illustrating a coaching method that takes into account the relationship type according to an embodiment of the present invention.

Referring to FIG. 3, in step 310, the coaching service server 200 acquires relationship data related to a relationship between a first object to be coached and a second object related to the first object.

In this case, the coaching may include providing coaching information related to the improvement of the relationship between the first object and the second object, and the second object may be an object of coaching when the coaching is executed.

For example, coaching is a gift recommendation, a first object is a person who makes a gift, and a second object is a person who will receive a gift. Thus, the coaching information may include information on a gift offering or behavior type recommendation that can be provided to the second object by the implementation of the first object.

Thus, the relationship formation relates to an interpersonal relationship between the first object and the second object, and the coaching may include coaching information for improving the interpersonal relationship between the first object and the second object.

At this time, the relationship data can be acquired and collected through various channels. For example, it may be collected through direct input of a user subscribed to a coaching service or through an application installed in a smart device.

In addition, the relationship data may be data in which the biometric data of the users are collected under a measurable environment.

For example, the relationship data may be data obtained by audition, may be data obtained by time, may include tactile data collected by a sensor or the like, biometric data such as motion / heart rate / brain waves of a pupil have.

For example, the heart rate information may be information that is loaded on a smartphone or measured through a peripheral (e. G., Earphone).

Heart rate data can be used to identify preferences through a general phenomenon of increased heart rate when looking at a particular product or to read emotions for a specific person. When talking to or interacting with a certain person, if the heart rate increases, the interpersonal relationship with the person can be interpreted as an initial stage or an inflection point, not a stabilization phase, and the interpersonal relationship may be a friendly relationship or a hostile relationship .

Therefore, in a friendly relationship, if the heart rate increases, a good situation can be deduced, and if it is a hostile relationship, it can be judged as a crisis situation. If the heart rate increases in a state where no specific relationship is established, it can be judged as a boundary / attention situation.

Relational data includes Social Networking Service (SNS) data associated with the first object and the second object, location information data associated with the first object and the second object, Message data exchanged between the first object and the second object, and call related data between the first object and the second object.

Thus, the relationship data can include a wide variety of data capable of analyzing human behavior.

The analysis of the relationship data may be analyzed periodically or non-periodically at specific points in time, and may be analyzed at a plurality of points of view considering sequential changes with time.

Various examples of channels through which relationship data is collected are as follows.

Relational data may be collected and analyzed at a particular point in time, and may be collected sequentially and repeatedly at multiple points in time.

For example, at least one of the search log and the cookie information using the device can be collected through the application installed in the smart device.

At this time, the search log may be collected by an application or operating system installed in the terminal, and may be, for example, an Internet surfing URL and search keyword data.

The URL and the search keyword may be collected via the cookie information recorded in the terminal or the computer, the smartphone application, or the operating system, and may be transmitted to the server.

Communication data such as a CDR (Call Detail Record) can also be collected as relational data, and the communication data includes a telephone frequency, a telephone usage period, a telephone time zone, etc., The contents of the conversation may also be collected.

In addition, the communication data may include a text message, an audio file, a moving picture file or tag, communication partner information, usage protocol information, and the like.

Relational data may include demographic data. For example, demographic data may include a variety of data such as an individual's gender, age, marital status, family composition information, current or desired occupation, current or desired income, and so on.

The communication data may include text data, images, images, etc., which are uploaded to or exchanged with an individual's SNS account. Data exchanged with these other people can be used to identify interpersonal types and stages.

Relational data may include moving image data, and in the case of image data obtained by CCTV, it may be assumed that it is possible for a lover relationship to be provided if the user moves to the other home at a later time.

Relational data may be collected from wearable devices such as smart glasses and smart watches that have been recently commercialized. In this case, various types of data may be collected. For example, it can include gaze movement data of a user watching text and images, and points for gazing at a photograph, a time of gazing, the number of times of gazing, and the like. Of course, since data collected from such wearable devices can only be acquired from a special environment or a limited user, the system can also determine the reliability or usability of the collected data in consideration of the nature of the collected data or equipment .

Eye movement data may be collected by a user who is watching text and images through equipment capable of acquiring eye movement data such as EyeTracking equipment or Google Glass.

For example, product preference can be grasped by utilizing the gaze fixed time and product information or image through the gaze movement data.

The relationship data basically includes position information data.

The location information data may be latitude or longitude data of the GPS, or may be location information obtained from the base station or the wireless access point. Further, in the case of a reader device installed in a specific place, position or place information may be recorded in advance in the system.

At this time, the location information data may be collected directly from the terminal or may be collected from the metadata of the photographs uploaded to the cloud storage. Also, the location information data may be collected together with the time information, and the travel route, the staying time, and the visit period may be collected.

The relationship data may include local information or place information corresponding to the location information. At this time, local information or place information may be deduced from the collected latitude and longitude information by the server, and specific building or mutual information such as map data of the navigation device may be directly collected.

For example, there is an amusement park in latitude A and longitude B, and it is presumed that there is a high possibility that the amusement park is classified as an amusement park and the relationship between lovers is frequent when two people who are not family relations are frequently visited together.

The collected data may include goods or service purchase data that can infer personal interests or hobbies.

The purchase data may include a purchase product name, a purchase time and time, a purchase situation, a purchase frequency, a purchase price, and the like.

Purchased product, purchase time / time, etc. can be used to identify the type of interpersonal relationship by grasping the person who was present and the person who was present, and can be used to grasp the consumption level or taste of the purchaser.

Also, depending on the type of the purchased item, it is possible to grasp the type of interpersonal relationship with the person who was accompanied with the purchase. For example, in the case of a lover, it is possible to deduce the stage of the lover's progress by the coffee shop settlement, the restaurant settlement, the movie settlement, the amusement park settlement, and the overseas travel settlement.

If other facial expression data can be collected, it may be possible to deduce the interpersonal stage with the other person by reading the expression of the person. In addition, the product's preference can be grasped by looking at the product.

When the voice data can be collected from the user terminal or the like, the emotion state of the person can be grasped by using the frequency and the amplitude of the voice call, the actual conversation, and the voice signal.

At this time, it is possible to extract information on the emotional state using natural language processing, speech to text function, text mining, and the like.

The voice data can be obtained through an acoustic input channel of a cellular phone, a PC, or a CCTV.

Other, if the skin conductivity data used in polygraphs can be collected, this data can be used to read human emotions.

Depending on the implementation environment, all collected data may be used to relate the relationship, and only some data may be used to relate the relationship. In addition, the collected data can be analyzed for all relationship types, and only the relationship types that can be derived from the collected data among all the relationship types can be inferred.

In step 320, the coaching service server 200 inferences the relationship type based on the collected relationship data.

In step 330, the coaching service server 200 determines the coaching method by considering the inferred relationship type.

In step 340, the coaching service server 200 may provide coaching information to the first object according to the determined coaching providing method.

For example, the timing of providing coaching information may be provided when a coaching target person desires, or when the coaching target person does not recognize, but it is determined that coaching is necessary.

At this time, the coaching information provision can be provided as text, image, moving picture, voice, and in the situation where predetermined condition is satisfied, there is a brain wave (there is a study example in which the body of another person is adjusted through brain wave interchange) Lt; RTI ID = 0.0 > and / or < / RTI >

At this time, the coaching service server 200 may classify the behavior of the object into "action" and "action occurrence status ".

Examples of behavior types

Visual behavior: gazing, gaze movement

Hearing: conversation, phone call

Virtual space act: message / image / video transmission / reception, SNS activity, online purchase, online site movement

Realistic space act: moving a place, buying offline

Example of behavior occurrence situation

Time situation Category: Private time zone, social time zone, private / social coexistence time zone, unconscious time zone, etc.

Real space situation Category: amusement park, nightlife, workplace, theater, etc.

Virtual space situation category: Shopping malls, movies, newspapers, search portals, etc. in various categories of online sites

It is possible to deduce interpersonal types and stages by analyzing the acquired data according to various combinations of the above (behavior type X behavior occurrence situation). The same data can be interpreted in various ways depending on each combination.

The type of interpersonal relationship can be classified as family, teacher, friend, lover, club member depending on the role. In addition, the type of interpersonal relationship can be classified into business relationship, seller-buyer relationship, and cooperation relationship depending on the purpose.

4 is a flowchart illustrating a coaching method considering a relationship type according to another embodiment of the present invention.

The embodiment shown in FIG. 4 may be performed after step 340 of providing the coaching information of FIG. Therefore, step 410 of FIG. 4 is the same as step 340 of FIG.

In step 420, the coaching service server 200 collects feedback information on the coaching providing result or feedback information on the coaching applying result.

In step 430, the coaching service server 200 may apply the collected feedback information to the coaching providing method.

Therefore, coaching can be performed repeatedly, not once at a time, and when the coaching subject's response or provided coaching is applied to the coaching applicant when the coaching is provided, the application result can be utilized to provide more appropriate coaching in the future have.

For example, the collection of data used to determine the response of the coaching subject is not collected on the terminal of the coaching subject and is collected through various channels. Similarly, the collection of data used to determine the response of the coaching applicant may not be collected solely for the terminal of the coaching applicant and may be collected over various channels.

5 is a flowchart illustrating a method of determining a coaching method according to an embodiment of the present invention.

Referring to FIG. 5, the step 330 of determining the coaching method of FIG. 3 includes a first procedure 331 for inferring a relationship forming step between the first object and the second object using the relation data, A second procedure 333 for detecting a situation requiring coaching of the second object, a third procedure 335 for determining coaching content, and a fourth procedure for grasping an acceptable coaching level of the first object (337).

Accordingly, the step 330 of determining the coaching method may include a step of grasping an event timing of the second object from the relational data, and a timing of providing coaching related to the second object to the first object, And a process for determining the current time.

At this time, the third procedure 335 may include a process of obtaining the preference and interest information of the second object, and a process of determining the content of the coaching considering the consumption level, preference and interest information.

In this case, the fourth procedure 337 may include a process of determining a consumption level of the first object.

At this time, the relationship type analogy and the relationship formation analogy can be inferred by taking into consideration the behavior type of the first object or the second object and information on the behavior occurrence status.

For example, the stage of interpersonal relationship means the stage of the relationship. In the business relationship, the progress stage can be classified by the increase of the reliability, and the friend relationship can classify the progress stage by the familiarity.

Classification by emotion can be categorized, for example, as liking, respect, love, hatred, dislike, jealousy.

Classification according to the tendency of interpersonal relationship formation stage can be classified as distance is getting closer, distance is getting away, depth is getting deeper, and depth is shallowing.

For example, the criterion for determining the level and extent of coaching is based on the preferred method of the coaching subject, the interpersonal relationship type of the coaching subject and the coaching subject, the interpersonal relationship stage of the coaching subject and the coaching subject, and the psychological or physical condition of the coaching subject Can be set.

For example, considering the fact that products that can be presented to a co-worker can be different from products that can be presented to a lover, and that products that can be presented at an early stage and those that can be presented at a mid- Can be determined.

Hereinafter, more specific examples will be described, for example, in the case where relationship formation is an interpersonal relationship and coaching is a gift presentation.

One way to recommend futures by coaching is to use a hybrid recommendation method that combines a content-based recommendation method and a collaborative filtering recommendation method.

At this time, the content-based recommendation method recommends the product based on the content of the product and the taste of the user, and the collaborative filtering recommendation method recommends a product that other users who are liked or interested by the user.

The coaching service server 200 may perform a recommendation of a gift by mixing the content-based recommendation method and the collaborative filtering recommendation method.

In order to recommend the object to be presented, the coaching service server 200 can grasp the type and level of interpersonal relationship of the user.

In addition, in order to recommend a time to present, the event time (birthday, anniversary, birth, etc.) of the characters in the interpersonal network can be grasped.

Also, in order to recommend a product to be presented, it is possible to grasp the product preference and interest of the people in the interpersonal network.

In addition, the level of consumption of the person to be presented can be grasped.

The coaching service server 200 grasps the timing of a gift, the preference of a gift recipient, and the consumption level of a person to be presented, and the interest of the gift recipient at an appropriate time based on the type of interpersonal relationship between the gift recipient and the provider, The user can recommend a product that meets the consumption level of the gift provider while belonging to the range.

In this case, the highest priority is placed on the consumption level of the gift supplier, the weight of one element is not satisfied, the timing of the gift, the preference of the gift recipient, You can also recommend a gift.

An analogy to the type of interpersonal relationship of a user, that is, a service subscriber, basically includes demographic information such as address, age, occupation, and sex of the user, location information that can be collected due to use of the terminal, Based on the character and photo information generated in < RTI ID = 0.0 >

For example, interpersonal types can include family relationships, friendships, heterosexual relationships, and peer relationships.

On the other hand, when information capable of accurately grasping the relationship is input or collected, the coaching service server 200 may determine the type of interpersonal relationship without going through a separate analogy process.

For example, if a carrier's family bundle discount, couple discount, family relationship certificate, graduation album, and user's address book are secured, the relationship between the two can be clearly grasped.

In addition, if the information collected from the user is input to the contact information, such as a father or a mother indicating the family relationship, the interpersonal relationship type can be clearly grasped.

As an example of demographic information, carrier subscription information may be utilized.

<An example of interpersonal relationship type of family relationship or family relationship>

In order to analyze the behavior of the user and deduce a family relationship and the like, usage log data of the user terminal, usage history information, and the like may be utilized.

At this time, the coaching service server 200 may classify the large category of the relationship first, and further refine the relationship to infer the interpersonal type. For example, if the address information of the users is the same, they can be inferred as "same address relationship" or "family relationship", and the type of relationship can be inferred such as father, mother, sister and younger brother through age information and gender information have.

Family relationships can be derived from the same address, appropriate age gap, gender composition, bedtime and sleeping time, time of arrival at the address and daytime activity area of the member. Addresses, age differences, gender composition can be identified through demographic information, bedtime and bedtime, time of entry into the address and time of activity of the member, and so on.

Recently, users of smartphones tend to use smartphones from the time of awakening to the moment they fall asleep. Therefore, sleep time and place, weather time and place can be collected through an application or an operating system that collects the usage history of the smartphone, and this information can be collected and used to derive the relationship type.

For example, if five persons are living in the same residence and five age groups are 70, 45, 43, 15, and 13 years old, the coaching service server 200 stores these five relationship types as family .

Also, if people in the same area sleep or wake up at similar times, they may be inferred as family relationships. Of course, at this time, the first use time and the last usage time of the terminal should be collected.

As another example, if a person is stationary at a certain point in time, for example, a birthday or wedding anniversary, a day on a holiday, a telephone call, and an SNS messenger, In other words, in the case of CDR data, if you contact a fixed point in time or retrieve a gift at the same time, you can acquire another backing basis. For example, if some of the people who are gathered in the same place or who are diagnosed as family members are getting to communicate with each other before the day of parental leave while maintaining the frequency of daily contact with the rest, The frequency of this is often the parent.

On the other hand, the daily life of a person is divided into two regions. Sleeping area and other activity areas. Here, the 'activity area' can be a job or a school. It is possible to define the activity area by analyzing the pattern such as visiting fixedly and staying for a certain time and going to and from certain regions. The school is the main activity area, the age is the student, and the age is the employee. The student is a child and the employee is likely to be a parent or a grown-up child. Employees can be divided into parents and older children.

You can also identify family relationships from images or texts that are uploaded to or exchanged with your SNS. The percentage included in the same image and the description of the image can be identified by the use of words that can analyze family relationships. If fixed images appear on several photographs at the same time, the possibility of a family relationship as a result of other data analysis becomes higher.

<Example of user's personal relationship formation step analogy>

Interpersonal relationships can be inferred through interaction frequency information, expression information, and location information.

The interpersonal relationship stage is arbitrarily divided into several stages, indicating that the relationship between the higher stages is strongly formed.

At this time, it can be deduced that the higher the frequency of contact, the higher the level.

The more time you stay together at the same location, the more you can guess.

The level of coaching may be determined by score by interpreting interpersonal stages. For example, if you recommend a gift by coaching, you can recommend a higher price product if it is inferred as a higher level.

The stage can be divided according to the point of contact which can be expressed by the contact day and the contact time. The higher the frequency of contact with private time or the time spent together, the higher the rank can be.

The higher the frequency of appearing together in the image, the higher the level can be.

As the location or background of the image is private, it can be classified into higher levels.

The smaller the proportion of formal expression through the expression used for contact, the higher the level may be.

The coaching service server 200 may assign predetermined scores to the exemplified classification criteria and estimate the maintenance, improvement, enhancement, and decline of the relationship formation step through the summed scores.

Coaching ranges and levels can be applied differently depending on the maintenance, enhancement, reinforcement, and decline of the relationship formation phase.

<Identification of event period of gift recipient>

The event information of the gift beneficiary subject to coaching by the user's execution of coaching may be disclosed information or collected personal event information. For example, the event time of the gift beneficiary may be information that is basically recorded in the personal information provided to the company such as a birthday or a wedding anniversary, and information disclosed in the SNS.

When the event time of the gift beneficiary is not clearly deduced, the coaching service server 200 may infer the timing of the event or the execution timing of the coaching through the related information.

For example, an event can be inferred by identifying a person's search pattern. For example, if the search keyword of the gift beneficiary is related to the birth, the change point of the search keyword can be grasped and the starting point of the birth can be guessed. As the frequency increases and the search keyword is associated with a particular product, it can be inferred that the time of birth is closer.

In addition, the event can be inferred by grasping a purchase pattern of a person. Depending on the product purchased, the date of birth can be deduced. The products purchased at the beginning of pregnancy and the products bought at the end of pregnancy are different. This information can be obtained through payment information, and can be obtained by identifying the location information of the person and the products to be sold at the corresponding location.

In situations where the phone number of the beneficiary is collected, the timing of the event may be inferred from the frequency of the contact. For example, if a gift recipient is identified by an obstetrician with an increased frequency of contact, the event can be inferred to be one month later, a few weeks later, and so on.

In addition, when the information on the hospital visit frequency and the visit period is obtained through the location information of the gift beneficiary, and information on the amount to be paid and the type of the inspections can be collected, the timing of the birth can be grasped and the coaching time can be determined.

<Suggested products for gift recipients>

Of course, the coaching service server 200 may perform coaching using a product recommendation method for an existing user himself / herself.

For example, it is possible to estimate a prize winner's preferred product when he / she searches for the prize winner, and to recommend the prize winner.

The coaching service server 200 can perform coaching by grasping the interests of the user who is a gift as a user. For example, through the usage information data (URL, search history), it is possible to check past favorite products through the user's current interests and purchase confirmation, and to further refine the products by age and gender.

For example, gift recipients will continue to visit the 'Prada' homepage, and if the past preferred products take up the bag, the gift beneficiary may be expected to receive the 'Prada bag' as a gift.

However, in the case of expensive products, it provides coaching information, provides information about the sale period of the high-priced products, and coaches "coaching information" about the special sale period of high- May be provided to the user.

The coaching service server 200 may provide coaching information through clustering of users. In order to classify users and clusters of similar users, individual user information data and previous purchasing history can be used. Identify the community to which the gift recipient belongs, and recommend a product satisfactory to the recipient based on the consumption characteristics in the community.

For example, if a gift recipient steadily searches for camping supplies, enters a related site, has a male gender, and is determined to have a spouse and children as a result of the relationship, it is included in the "family camping" When searching for a "hammer", it is possible to recommend a preferred camping hammer in the community to the coaching person.

<Inferred level and range of recommendation of gift provider>

An example of recommendation acceptance level of the gift provider is the consumption level, and examples of the recommendation acceptance range may be whether it is a private gift gift or a public gift gift.

The recommendation acceptance level and range can be grasped based on the purchase history of the gift provider and the type of relationship with the gift recipient.

It is also possible to subdivide the purchasing history of the user into each relationship type, or to subdivide the recommendation acceptance level and range for each relationship formation step.

In addition, trends in recommendation acceptance levels and ranges may be identified and inferred to be acceptable levels and ranges at this time.

6A is a diagram illustrating a configuration of an apparatus for providing coaching information through identification of an interpersonal relationship type according to an embodiment of the present invention.

At this time, the coaching information providing system may be the coaching system shown in FIG.

6A, the coaching information providing apparatus 600 includes a database 610, a relationship data acquiring unit 620, a record generating unit 630, and an interpersonal relationship type determining unit 640 do.

In the database 610, various data generated in the coaching information providing system may be stored in a database.

For example, the database 610 may include an interpersonal type database in which all types of interpersonal relations to be identified through the coaching information providing system are defined.

Types of interpersonal relationships can include, for example, family members, friends, lovers, co-workers, classmates, fellowship members, and priesthood partners. More specifically, in the case of a family member, a specific type of family member such as "father and daughter" may be defined.

The database 610 may also store a data field table for defining interpersonal relationship types.

The relationship data acquiring unit 620 acquires relationship data related to the relationship between the first object and the second object related to the first object.

The record generation unit 630 generates a record for each item of the data field table from the relational data.

The interpersonal relationship type determination unit 640 determines an interpersonal relationship type between the first object and the second object based on the generated record.

The interpersonal relationship type determination unit 640 calculates an index value for each interpersonal relationship type every time a record is generated, calculates an average of the index values for records of a certain period to be analyzed, The trend of the interpersonal relationship step can be calculated by comparing the arithmetic mean of the exponent value and the exponential weighted average of the exponent value.

The interpersonal relationship type determination unit 640 determines whether or not the newly generated record is matched with the record patterns derived for each interpersonal type each time a record is newly generated, Based on whether or not the maximum pattern number value is greater than or equal to a preset threshold value, identifying an interpersonal relationship type having a maximum pattern number value having a largest value of the number of times belonging to each interpersonal type pattern, And determine an interpersonal type of the object and the second object.

The interpersonal relationship type determination unit 640 applies a record to the pre-established interpersonal relationship classification model every time a record is generated, and determines the interpersonal type of the first object and the second object based on the error rate of the classification model You can decide.

At this time, the interpersonal relationship type determination unit 640 may include an apparatus for calculating and determining an index for each relationship type, as shown in FIG. 6B.

The apparatus for calculating and determining an index for each relationship type may calculate an index value for each interpersonal relationship type in which the interpersonal relationship type is defined, whenever the record is generated.

The apparatus for calculating and determining the indices according to the relationship type includes a process of calculating an association index between the occurrence time of the relationship type and the corresponding relationship type, a process of calculating the association index between the occurrence place of the relationship type and the corresponding relationship type, And the relationship index is calculated, and the association index between the subject and the relationship type is calculated. The importance of each time, place, situation, subject, and importance of each attribute are applied as weights And an association index may be calculated.

6B, an apparatus for calculating and determining an index for each relationship type includes an action / activity occurrence time data verifying unit 641 for confirming an action or activity occurrence time data, An activity / activity occurrence status data confirmation unit 643 for confirming the activity / activity occurrence status data acquired by the sensor, an activity / activity occurrence status data confirmation unit 643 for confirming data of the activity or activity target person An activity / activity data confirming unit 644 for confirming an activity or an activity content, an activity / activity data confirming unit 645 for confirming an activity / activity content, an activity / activity result data confirming unit 646 for confirming an activity or an activity result, A relationship type determination unit 647 that calculates an index for each relationship type from the relationship type determination unit 647, and a relationship type determination unit 648 that determines a relationship type in consideration of the relationship type index. .

For example, the act may be to provide a gift, and the result of the act may be the response of the recipient (joy, surprise, reply to the thank-you message).

Here, an action refers to a single action of an object, and an action may be an action that a plurality of objects together. Therefore, data on activity / activity occurrence time, activity / activity occurrence location, activity / activity occurrence status, activity / activity result data, etc. may include data on actors as well as actors such as actors. For example, when the first object calls the second object, the place of the act of making a phone call may be the office of the first object and the second object may be at home.

Although not shown in FIG. 6A, the coaching information providing apparatus 600 through interpersonal type grasping can determine a coaching method for the first object or the second object in consideration of the determined interpersonal relationship type, .

Hereinafter, a concrete method for grasping interpersonal relationship type will be described.

FIG. 7A is a flowchart illustrating a method of providing coaching information through identification of an interpersonal relationship type in an interpersonal coaching system according to an exemplary embodiment of the present invention.

The method shown in FIG. 7A can be performed by the coaching information providing apparatus shown in FIG. 6 or the relationship type inference unit 230 in FIG.

Referring to FIG. 7A, in step 710, the coaching information providing apparatus maintains an interpersonal relationship type database in which an interpersonal relationship type is defined.

In step 720, the coaching information providing apparatus defines and maintains a data field table for defining an interpersonal relationship type.

Here, "data field" means an attribute of data.

For example, the data field included in the data field table may be composed of 36 fields as shown in [Table 1].

[Table 1]

In Table 1, "event subject" refers to the subject who generated the event. Here, "event" is a term used in the database, which means an action of an object.

For example, if you call a person named A from a phone terminal of person A, the event subject is person A.

As described above, each item of the data field table includes an event subject that defines a subject that has generated an event related to the relationship formation, an object of the event, a synchronization of the event, information on a time when the event occurred, The presence information of the event, the demographic information of the event related person, the status information of the event related person, the action method inferred from the relationship data, the analogy from the relationship data, Current or future issue information, and personal information of the first object or the second object.

Data fields can be expanded or added in the process of collecting and analyzing relationship data.

In step 730, the coaching information providing apparatus collects relationship data related to the relationship between the first object and the second object. Relational data may be acquired and collected over various channels.

In step 740, the coin information providing apparatus generates a record for each item of the data field table from the relational data.

At this time, the record means information that is recorded in the database after extracting a value matching the data field from the relationship data. Thus, the record may only be generated from the values of the data fields obtainable from the relationship data.

The values that are matched to the data fields in the relational data can all be represented by the attribute values of the records. For example, the attribute value of a record can be defined as a number for displaying the attribute of the record.

[Table 2] shows an example of generation of records per data field.

In the example of [Table 2], Cheol-su and Ye-hee are captains, and Cheol-su and Dong-su are rich. He works at the company between 9: 00 ~ 20: 00 and Yeon Hee works at the mart between 4: 30 ~ 7: 30. He is a retired father. She greeted her on Mother's Day and gave thanks for using KakaoTalk and made voice calls using KakaoTalk's Boystock feature to discuss her family's travel plans over the coming weekend.

In step 750, the coaching information providing system determines an interpersonal relationship type between the first object and the second object based on the generated record.

On the other hand, the interpersonal coaching system can perform a procedure of selecting relationship data for the first object and the second object, as shown in FIG. 7B, when confirming or collecting the relational data.

Referring to FIG. 7B, when data exists, it is determined in step 701 whether the contents of the data are related to the object (the first object or the second object).

If the contents of the data are related to the object, the relation data can be determined in step 707. Otherwise, it is determined in step 703 whether the data generation process is related to the object.

If the data generation process is related to the object, it can be determined as the relational data. Otherwise, it can be determined in step 705 whether the data generation space is related to both objects.

If the data creation space is related to both objects, it can be determined as relational data in step 707. Otherwise, it can be determined in step 709 that relational data is not for two objects.

Although not shown in FIG. 7A, the coaching information providing method may further include determining a method of providing coaching for the first object or the second object in consideration of the determined interpersonal relationship type.

At this time, the coaching information includes information on a gift providing or behavior kind recommendation that can be provided to the second object by the implementation of the first object.

In this case, the step of determining a coaching providing method may include a first procedure for inferring a relationship forming step between the first object and the second object using the relational data, a step of detecting a situation requiring coaching of the second object, A third procedure for determining coaching content, and a fourth procedure for grasping an acceptable level of coaching of the first object.

At this time, the fourth procedure checks (761) the behavior history data of the first object and the second object included in the relationship data as shown in FIG. 7C, and classifies the behavior history by the relation type from the behavior history data A process of dividing the behavior history by the relationship formation step 763, a process of dividing the behavior history by the relation trend 764, a process of classifying the behavior type (765) extracting a behavior level pattern for each combination of a relationship type, a relationship formation step and a relationship trend, a step (766-1) for extracting a behavior level pattern for each combination of the relationship type, the relationship formation step, A process of extracting a provision route pattern 766-2, or a process of extracting a behavior pattern (type) pattern 767 for each relationship type / relation formation step / relation tendency (step 767) can do.

8 and 9 are flowcharts for explaining an interpersonal relationship type determination method according to an embodiment of the present invention.

The method shown in Figs. 8 and 9 may be performed in step 750 of Fig. 7A, respectively.

Referring to FIG. 8, in step 851, each time a record is generated, the coaching information providing apparatus may calculate an index value for each interpersonal relationship type.

The index value can be calculated by interpersonal type, and can include, for example, a family relationship index, a friend relationship index, a love relationship index, a work colleague relationship index, a study relationship index, a social group membership index,

An index value may be assigned to one record, or an index value may be given in consideration of the relationship between the records. For example, if the 'gender (person)' and '34) gender (gift beneficiary)' of both '1) event subject' and '2) And 'peer relationship index' can be given a high value, and 'lover relation index' can be given the lowest value.

The exponent value may be set to be calculated from the attribute values of the record and the demographic data values, or the exponent value may be set in advance for each attribute of the record in the system.

If an expression for calculating the interpolation index value is set, the index value may be calculated by extracting the corresponding item of the calculation expression from the records generated for a predetermined period of time.

The calculation of the exponent value can be calculated by considering the data such as the occurrence time and place, the occurrence situation and the subject of the behavior, and the interpersonal characteristics of the behaviors that can have different meanings according to interpersonal relations.

In addition, the calculation of the exponent value may be performed by weighting the importance of the specific data field or by considering the weight of the data field by category. For example, it is also possible to assign importance to the time of occurrence, to divide the case of the phone call and the case of the message transmission, and to divide the weight according to the category and reflect the result in the calculation.

In step 853, the coaching information providing apparatus calculates an average of index values calculated for interpersonal relationship types for records of a predetermined period to be analyzed. The average of the index values by interpersonal type can be utilized variously.

For example, the average of the index values by type of interpersonal relationship can be expressed as [Table 3].

[Table 3]

As shown in [Table 3], the average of the index values by type of interpersonal relationship can be calculated in various ways such as arithmetic mean and exponential weighted average. In addition, any statistical value that can represent the exponent value can be utilized.

If the arithmetic mean of the exponent value and the exponentially weighted average of the exponent value are calculated, a meaningful analysis can be performed by comparing the arithmetic mean of the exponent value and the exponentially weighted average of the exponent value.

In step 855, the coaching information providing apparatus may calculate the trend of the interpersonal relationship level by comparing the arithmetic mean of the exponent value and the exponential weighted average of the exponent value.

For example, if the exponentially weighted average of the lover relationship index is greater than the output average, this can be interpreted as a progress of the lover relationship. The exponentially weighted average is characterized by giving more weight to recent exponential values.

Thus, in the state of interpersonal relationship type, the interpersonal relationship can be determined through analysis of the exponent value, postponement, regression, etc., and the contents of coaching can be changed according to the trend of interpersonal level.

On the other hand, the ratio of the individual index value averages to the sum of the index value averages for all interpersonal types is expressed as a percentage, which can be referred to as an "index percentage ".

In Table 3, the sum of the exponential averages is 12.5 for the arithmetic mean, the arithmetic mean of the family relations index is 4.5 / 12.5, and the exponent percentage is 36%.

At this time, the process of determining the interpersonal relationship type may be such that each time a record is generated, an index value is calculated for each interpersonal relationship type, and when the sum of the index values during a certain period of time to be analyzed is a maximum Identifying an interpersonal relationship type having an exponent value and determining an interpersonal type of the first object and the second object based on whether the sum of the maximum exponent values is equal to or greater than a predetermined threshold.

If the categories of interpersonal types to be identified are defined mutually exclusively, if the sum of the maximum exponent values is greater than or equal to the specified threshold value, the corresponding interpersonal relationship type is defined as an interpersonal relationship type for the records of two persons You can decide.

If there is no interpersonal relationship type with an exponent value that exceeds the specified threshold, the interpersonal relationship type presentation is suspended and the decision may be held until a significant change is made in the sum of the exponent values by obtaining additional data.

Accordingly, in determining the interpersonal type of the first object and the second object based on whether the maximum exponent value is greater than or equal to a preset threshold value, if the sum of the maximum exponent values is smaller than a predetermined threshold value, Determining whether a plurality of interpersonal relationship types are mutually exclusive when the sum of the index values is more than or equal to the threshold value; determining whether a plurality of interpersonal relationship types are mutually exclusive; And the interpersonal relationship type of the second object.

If the categories of interpersonal types to be identified are not defined mutually exclusive, then all interpersonal type (s) corresponding to the index percent (s) above the threshold must be interpolated for records of two persons Type. &Lt; / RTI &gt;

At this time, the threshold value to be used may be determined in consideration of the number of interpersonal relations to be considered.

If there is no interpersonal relationship type with an exponent percentage that exceeds the specified threshold, the interpersonal relationship type presentation is suspended and the decision can be held until a significant change is made in the exponential percentage by acquiring additional data. If interpersonal relationships are more than two types of coaching, the coaching content may be provided for each inferred type.

Therefore, in determining the interpersonal type of the first object and the second object based on whether the sum of the maximum exponent values is equal to or greater than a preset threshold value, the interpersonal type in which the sum of the exponent values is greater than or equal to the threshold value And a plurality of interpersonal types having a plurality of the index values equal to or greater than the threshold value are mutually exclusive, an interpersonal relationship type having a sum of the maximum exponent values is determined as an interpersonal type of the first object and the second object .

In one embodiment, the step of determining an interpersonal relationship type comprises the steps of: calculating an index value for each interpersonal relationship type in which a plurality of interpersonal relationship types are mutually exclusive defined each time the record is generated; Determining an interpersonal relationship type having a sum of the maximum exponent values of the sum of the exponent values for a predetermined period of time as a target, and determining whether the sum of the maximum exponent values is greater than or equal to a predetermined threshold value And determining an interpersonal type of the object and the second object.

If the sum of the maximum exponent values is smaller than a predetermined threshold value, the interpreter type determination is suspended. If the sum of the maximum exponent values is greater than or equal to the threshold value, 1 &lt; / RTI &gt; object and the second object.

In one embodiment, the step of determining an interpersonal relationship type comprises: calculating an exponent percentage for each interpersonal relationship type in which a plurality of interpersonal relationship types are not mutually exclusive defined each time the record is generated; Identifying an interpersonal relationship type having a maximum exponent percentage value with the largest exponent percentage value for a predetermined period of time, and determining whether the first exponent percentage value is greater than or equal to a predetermined threshold value, And determining an interpersonal type of the object.

If the maximum exponent percentage value is smaller than the predetermined threshold value, the interpersonal relationship type determination is suspended. If the interpersonal relationship type having the exponent percentage value equal to or larger than the threshold value is a plurality of interpersonal relationship types, And the interpersonal relationship type of the object and the second object.

FIG. 9 can be applied to a case where record patterns can be derived for each type of interpersonal relationship through accumulated records for each type of interpersonal relationship.

In order to derive the record patterns, it is possible to divide the people into several groups using Clustering Analysis based on the similarity of the attributes of people who want to understand interpersonal type. The attributes used at this time are effective properties for interpersonal type. Valid attributes can be derived through literature studies or statistical techniques. Clustering analysis can be carried out according to interpersonal type to be identified, and clusters can be formed by interpersonal type. In other words, people can be divided into several clusters based on similarity only in terms of family relations, and people can be divided into several clusters on the basis of similarity only in terms of peer relationships. Or several types of interpersonal relationships, regardless of interpersonal relationship type.

For each of the generated clusters, patterns that predominate in a particular interpersonal type can be derived.

For example, let's say you want to understand family relationships, friendships, love relationships, work colleagues, coworkers, friendship, and priest relationships. In terms of family relations, we conducted a cluster analysis on the subjects, and five clusters were formed. Let A, B, C, D, and E be the clusters. For example, the characteristics of the derived A community may be as follows. The two people to be identified are men and women, men are between 25 and 45 years old, women are between 22 and 40 years old, and men and women are within 30 km away from each other. An example of a pattern that may frequently appear as a family relationship in the A community is as follows.

Men who are subject to interpersonal relationships belonging to group A mainly show the following patterns of life. A man belonging to a group A is between 6:00 and 7:00, between Monday and Friday, at a time belonging to the man's private time category, at home, to the man, the house is a private real space, 100% of the time during the stay in the house is to check the KakaoTalk for confirmation of news or messages to him, KakaoTalk is a private internet space, and this guy is in the virtual space, KakaoTalk And the time is less than 10 minutes. This man's working day is weekly, and his age is between 25 and 45 years old. These patterns can be derived through literature studies. Also, if you have enough records to be generated among people whose interpersonal types are already known, you can derive them through a classification analysis. For example, a path from a Decision Tree to a leaf node can form a pattern.

Referring again to FIG. 9, in step 951, each time a record is newly generated, the coaching information providing apparatus determines whether or not the newly generated record is matched with the record patterns derived for each interpersonal type.

In step 953, the coin information providing apparatus checks the interpersonal type having the maximum pattern count value having the largest number of times belonging to the patterns for each interpersonal type, with respect to the records generated for a predetermined period to be analyzed.

In step 955, the coaching information providing apparatus may determine the interpersonal type of the first object and the second object based on whether the maximum pattern number value is equal to or greater than a preset threshold value.

The more interpersonal types that have a large number of patterns to which the records belong, the more likely they are the types of interpersonal relationships they are forming. Therefore, the number of records belonging to the pattern of each interpersonal type can be counted for the type of the alternative relationship and used for the interpersonal type analogy with respect to records of a certain period to be analyzed.

Regardless of the type of interpersonal relationship, the sum of the number of times the records belonging to a certain period of time, which are analyzed for a certain period of time, belongs to all the patterns can be calculated, and this can be referred to as the sum of the number of times. The sum of the number of times the records of a certain period of analysis to be analyzed belongs to patterns belonging to each interpersonal type is calculated and is called a pattern frequency value. For each interpersonal type, (pattern frequency value) / (total frequency value) * 100 is calculated, and this can be referred to as a pattern frequency percentage for the interpersonal type.

If the categories of interpersonal types to be identified are mutually exclusive, if the maximum number of pattern values is greater than the specified threshold value, the corresponding interpersonal relationship type is determined to be the interpersonal relationship type for the records of the two persons .

If there is no interpersonal relationship type with a pattern count value that exceeds the specified threshold, the interpersonal relationship type presentation is suspended and the decision can be suspended until additional data is acquired and a significant change in the pattern frequency value occurs.

Therefore, when determining the interpersonal type of the first object and the second object based on whether the maximum pattern number value is greater than or equal to a preset threshold value, if the maximum pattern number value is less than the predetermined threshold value, Determining whether a plurality of interpersonal relationship types are mutually exclusive if the plurality of interpersonal relationship types having the pattern number value equal to or greater than the threshold value are determined to be mutually exclusive; And the interpersonal relationship type of the second object.

If the categories of interpersonal types to be identified are not defined mutually exclusive, then all interpersonal type (s) corresponding to the number of times pattern percentage (s) above the threshold value are interpolated for the records of the two persons You can decide by relationship type.

The threshold used is determined taking into account the number of interpersonal relationships considered. If there is no interpersonal type with a pattern count percentage greater than the specified threshold, the interpersonal type presentation is suspended and the judgment can be suspended until additional data is acquired and a significant change in the pattern frequency percentage occurs.

Accordingly, in determining the interpersonal type of the first object and the second object based on whether the maximum pattern number value is greater than or equal to a preset threshold value, the interpersonal type having the maximum pattern number value equal to or greater than the threshold value And the plurality of interpersonal types having a plurality of pattern count values equal to or greater than the threshold value are mutually exclusive, the interpersonal relationship type having the maximum pattern count value can be determined as the interpersonal type of the first object and the second object .

In one embodiment, the step of determining an interpersonal relationship type includes: generating a record pattern that is generated for each type of interpersonal relationship in which a plurality of interpersonal types are mutually exclusive defined, Determining an interpersonal type having a maximum number of pattern values having the largest number of times belonging to the patterns for each interpersonal type, with respect to the records generated for a predetermined period of time to be analyzed; And determining an interpersonal type of the first object and the second object based on whether the maximum pattern number value is greater than or equal to a predetermined threshold value.

If the maximum pattern count value is less than the predetermined threshold value, the interpersonal relationship type determination is suspended. If the maximum pattern number value is greater than or equal to the threshold value, the interpersonal relationship type having the maximum pattern count value is stored in the first object And the interpersonal relationship type of the second object.

In one embodiment, the step of determining the interpersonal relationship type may further include, for each new record generated, the newly generated record is a record of a plurality of interpersonal relationship types that are derived for each type of interpersonal relationship Determining an interpersonal relationship type having a pattern frequency percentage value having a largest pattern frequency percentage value for records generated for a predetermined period of time to be analyzed; Determining a type of interpersonal relationship between the first object and the second object based on whether the frequency percentage value is equal to or greater than a predetermined threshold value.

In this case, if the percentage of the maximum number of patterns is less than a predetermined threshold value, the determination of the interpersonal relationship type is suspended. When the plurality of interpersonal types having the pattern number percentage value equal to or greater than the threshold value are plural, And can be determined as an interpersonal relationship type between the first object and the second object.

Similar to the method described in Fig. 9, when data related to users with interpersonal relationship types are sufficiently secured, an interpersonal relationship classification model is established, and an interpersonal relationship classification model . &Lt; / RTI &gt;

For example, each time the record is generated, the record may be applied to a pre-established interpersonal relationship classification model to determine the interpersonal type of the first object and the second object based on the error rate of the classification model .

At this time, the interpersonal relationship classification model may include a classification model for interpersonal relationship type and a detailed classification model for interpersonal relationship type, which are established based on behavioral characteristics by interpersonal relationship type.

Here, the classification model for interpersonal relationship type is a top concept model for classifying interpersonal relationships such as "family relationship" and "friend relationship", and the detailed classification model for interpersonal relationship type is "weekend couple", " Relationship "," hat ", and" rich ".

Hereinafter, a description will be made of a process of establishing a detailed classification model for interpersonal relationship type and interpersonal relationship type, and an exemplary method for classifying interpersonal relationship type using the same.

<Establishment and Utilization of Classification Model for Interpersonal Relations Type>

The analysis subjects who know interpersonal relationship type can be divided into several groups with similar behavior characteristics for each interpersonal relationship type. At this time, Clustering Analysis can be used for group classification, and record attributes can be used.

 For each group, a model can be established to classify interpersonal relationships.

At this time, the classification value of the model can be "family relationship", "friend relationship", "lover relationship", "peer relationship", "friendship group member relationship", " For each group, classify all interpersonal types as one model.

Whenever a new record of the objects to be interpolated is generated, the classification value of the interpersonal relationship type and the error rate can be obtained by applying a model to each interpersonal relationship type. Using the classification value and the error rate obtained after applying the model, the type of interpersonal relationship between the specific gift provider and the gift recipient can be grasped. The lower the error rate, the higher the likelihood that the two individuals are interpersonal.

Only when the error rate is equal to or less than the specified threshold value, the classification value corresponding thereto is judged to be significant.

All the classification values having an error rate equal to or less than a specified threshold value are searched for records of a predetermined period to be analyzed.

Using the frequency of the interpersonal relationship type and the error rate of the obtained classification values, the type of interpersonal relationship between the specific gift provider and the gift beneficiary can be grasped. For example, an interpersonal type corresponding to a classification value having a maximum frequency can be presented as an analogy value.

The error rate adjusted frequency, which is the frequency adjusted by the error rate, can be used. For example, in a single occurrence, the frequency calculation may include (1-error rate). If the error rate is 0%, 1-0 = 1. If the error rate is 100%, 1-1 = 0. For example, let's say you have a family relationship check with a 10% error rate, a family relationship check with a 5% error rate, and a family relationship check with a 1% error rate. Let's also assume that the peer relationship determination with 1% error rate, the peer relationship determination with 1% error rate, and the peer relationship determination with 1% error rate are confirmed from the records. In this case, the number of family relationship judgments and the number of judges of peer relations are 3 as the unadjusted frequency. However, as the error rate and the adjusted frequency, the number of family relationship determinations was (1-0.1) + (1-0.05) + (1-0.01) = 2.84, (1-0.01) = 2.97. Thus, based on the error rate and the adjusted frequency, the two are more likely to be coworkers.

Percentage represents the percentage of the frequency of each type relative to the sum of the frequencies for all relationship types. Let's say this is a percentage. Here, the frequency may be the frequency adjusted by the error rate. Here, the total of the frequencies refers to the number of classification values when all the classification values having an error rate equal to or less than a specified threshold value are obtained for records of a predetermined period to be analyzed. And the frequency by type means the number of classification values corresponding to the same interpersonal type. In other words, the sum of frequencies for each type of interpersonal relationship is the sum of frequencies.

If the categories of interpersonal types to be identified are defined mutually exclusive, if the maximum frequency is greater than the specified threshold value, the corresponding interpersonal relationship type is presented as an interpersonal relationship type for the records of two persons.

If there is no interpersonal relationship type with a frequency greater than or equal to the specified threshold, the interpersonal relationship type presentation is suspended and the decision can be held until additional data is acquired and a significant change in frequency occurs.

If the categories of interpersonal types to be identified are not defined mutually exclusive, then all interpersonal type (s) corresponding to the frequency percent (s) above the specified threshold value are interpolated for the records of the two persons Relationship type.

The threshold used is determined taking into account the number of interpersonal relationships considered. If there is no interpersonal type with a frequency percentage greater than or equal to the specified threshold, the interpersonal type presentation is suspended and the decision can be held until additional data is acquired and a significant change is made in the frequency percent.

Clustering analysis and classification analysis can be performed again periodically using newly acquired data to establish a model.

<Establishment and utilization of detailed classification model for interpersonal type>

The analysis subjects who know interpersonal relationship type can be divided into several groups with similar behavior characteristics for each interpersonal relationship type. At this time, Clustering Analysis can be used for group classification, and record attributes can be used.

A model can be established that classifies the interpersonal relationship type for each of the groups of analyzed persons divided for each interpersonal relationship type. For example, the number of all the subjects to be analyzed can be divided into five groups (A, B, C, D, and E groups) based on similarity criteria based on behaviors seen in family relationships of 10 million.

For each of the five groups, a model that classifies whether or not they are family relationships can be established using classification analysis. That is, five classification models A_M, B_M, C_M, D_M, and E_M can be generated.

If the behaviors of the gift provider and the gift beneficiary who want to understand the interpersonal relationship are closest to the A group, the A_M is used to classify whether the relationship is family relationship or not. Each time you apply A_M to each record, it will be classified as family relationship or non-family relationship. The error rate, which can be calculated at the time of model establishment, is also known.

For other types of interpersonal relationships, models can be established and the model of each record can be applied to obtain the classification value and error rate of the interpersonal type.

Whenever a new record of objects to be interpolated is created, the type of interpersonal relationship between the specific gift provider and the recipient of the gift can be grasped by using the classification value and the error rate obtained after applying the model for each interpersonal type . The lower the error rate, the higher the likelihood that the two individuals are interpersonal.

The error rate average of each relationship type is calculated for records of a certain period to be analyzed. The error rate average can be calculated in several ways, such as arithmetic mean and exponential weighted average, and any statistical quantity that can represent the error rate to be calculated can be utilized. In addition, both the arithmetic average and the exponentially weighted average are obtained and compared to each other, so that the trend of the interpersonal stage can be grasped. If the exponentially weighted average of error rates of lover relations is smaller than the arithmetic mean, it can be considered that there is progress in lover relationship. And the exponentially weighted average gives more weight to the recent error rate. Depending on the progress of interpersonal relationship, backstroke, and backwardness, coaching contents may be different.

Percentage of the percentage of individual error rate averages over the sum of error rate averages for all relationship types. This can be referred to as an error rate percentage.

If the categories of interpersonal types to be identified are defined mutually exclusive, if the minimum error rate is below the specified threshold value, the corresponding interpersonal relationship type can be determined as the interpersonal relationship type for the records of the two persons .

If there is no interpersonal relationship type with an error rate less than or equal to the specified threshold, the interpersonal relationship type presentation is suspended and the decision can be suspended until additional data is acquired and a significant change in error rate occurs.

If the categories of interpersonal types to be identified are not defined mutually exclusive, then all interpersonal type (s) corresponding to the error rate percent (s) below the specified threshold value are interpolated for the records of the two persons You can decide by relationship type.

If there is no interpersonal relationship type with an error rate percentage that is below the specified threshold, the interpersonal relationship type presentation is suspended and the decision can be suspended until additional data is obtained and a significant change is made in the percent error rate.

Clustering analysis and classification analysis can be performed again periodically using newly acquired data to establish a model.

Thus, in one embodiment, the step of determining the interpersonal relationship type comprises applying the record to a single model, which is configured to classify all interpersonal relationship types each time the record is generated, Calculating an error rate for the corresponding type among the externally defined interpersonal types; and calculating a maximum error rate adjustment frequency with a largest error rate adjusted frequency based on the error rate for a predetermined period of time, And determining an interpersonal type of the first object and the second object based on whether the maximum error rate adjustment frequency value is equal to or greater than a predetermined threshold value .

If the maximum error rate adjustment frequency value is smaller than the preset threshold value, the interpersonal relationship type determination is suspended. If the maximum error rate adjustment frequency value is greater than or equal to the threshold value, the interpersonal relationship type having the maximum error rate adjustment frequency value 1 &lt; / RTI &gt; object and the second object.

In one embodiment, the step of determining the interpersonal relationship type comprises: applying the record to one model, which is configured to classify all interpersonal relationship types each time the record is created, so that the plurality of interpersonal types are mutually exclusive Calculating an error rate for the corresponding one of the undefined interpersonal types; and identifying an interpersonal type having a maximum frequency percentage value with the highest frequency percentage value based on the error rate for a certain period of time to be analyzed step; And determining an interpersonal type of the first object and the second object based on whether the maximum frequency percentage value is greater than or equal to a predetermined threshold value.

If the maximum frequency percentage value is smaller than the predetermined threshold value, the determination of the interpersonal relationship type is suspended. If the interpersonal relationship type having the frequency percentage value of the threshold value or more is a plurality of the plurality of interpersonal types, And the interpersonal relationship type of the object and the second object.

In one embodiment, the step of determining the interpersonal type comprises: identifying a cluster that is most similar to object 1 and object 2 for pre-defined pseudo-object clusters, and generating, for each of the records, Calculating an error rate for a corresponding type of interpersonal relationship types in which a plurality of interpersonal relationship types are mutually exclusive defined by applying the record to a previously established classification model, Determining an interpersonal relationship type having a maximum error rate adjustment frequency value in which a value of an error rate adjusted frequency is largest and determining whether the maximum error rate adjustment frequency value is greater than or equal to a preset threshold value, And determining an interpersonal type of the second object.

If the maximum error rate adjustment frequency value is smaller than the preset threshold value, the interpersonal relationship type determination is suspended. If the maximum error rate adjustment frequency value is greater than or equal to the threshold value, the interpersonal relationship type having the maximum error rate adjustment frequency value 1 &lt; / RTI &gt; object and the second object.

In one embodiment, the step of determining the interpersonal type comprises: identifying a cluster that is most similar to object 1 and object 2 for pre-defined pseudo-object clusters, and generating, for each of the records, Calculating an error rate for a corresponding type of interpersonal relationship types in which a plurality of interpersonal relationship types are not defined mutually exclusive by applying the record to a previously established classification model; Identifying an interpersonal relationship type having a maximum frequency percent value with a largest value of the frequency percent based; And determining an interpersonal type of the first object and the second object based on whether the maximum frequency percentage value is greater than or equal to a predetermined threshold value.

If the maximum frequency percentage value is smaller than the predetermined threshold value, the determination of the interpersonal relationship type is suspended. If the interpersonal relationship type having the frequency percentage value of the threshold value or more is a plurality of the plurality of interpersonal types, And the interpersonal relationship type of the object and the second object.

In one embodiment, the step of determining the interpersonal relationship type comprises: identifying clusters most similar to the object 1 and the object 2, one for each of the interpersonal relationship types, for the similar object clusters by the predefined interpersonal type, Calculating an error rate for each type of interpersonal relationship in which a plurality of interpersonal types are defined mutually exclusive by applying the record to each of the corresponding interpersonal type classification models established for each of the most similar clusters Determining an interpersonal relationship type having a minimum average error rate value having a smallest average of the error rates for a predetermined period of time to be analyzed and determining whether the minimum average error rate value is less than a preset threshold value, And determining an interpersonal type of the first object and the second object.

If the minimum average error rate value is greater than a predetermined threshold value, the interpersonal relationship type determination is suspended. If the minimum average error rate value is less than or equal to the threshold value, the interpersonal relationship type having the minimum average error rate value, It can be determined as the interpersonal type of the second object.

In one embodiment, the step of determining the interpersonal relationship type comprises: identifying clusters most similar to the object 1 and the object 2, one for each of the interpersonal relationship types, for the similar object clusters by the predefined interpersonal type, Calculating the error rate for each type of interpersonal relationship in which a plurality of interpersonal relationship types are not defined mutually exclusive by applying the record to each of the corresponding interpersonal type classification models established for each of the most similar clusters, Determining an interpersonal relationship type having a minimum error rate percentage value having a smallest value of an error rate percentage based on the error rate for a predetermined period of time, and determining whether the first error rate percentage value is less than a predetermined threshold value And determining an interpersonal type of the object and the second object .

In this case, if the minimum error rate percentage value is larger than the predetermined threshold value, the interpersonal relationship type determination is suspended, and if the plurality of interpersonal relationship types having the error rate percentage value equal to or less than the threshold value are plural, And the interpersonal relationship type of the second object.

The apparatus described above may be implemented as a hardware component, a software component, and / or a combination of hardware components and software components. For example, the apparatus and components described in the embodiments may be implemented within a computer system, such as, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable array (FPA) A programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions. The processing device may execute an operating system (OS) and one or more software applications running on the operating system. The processing device may also access, store, manipulate, process, and generate data in response to execution of the software. For ease of understanding, the processing apparatus may be described as being used singly, but those skilled in the art will recognize that the processing apparatus may have a plurality of processing elements and / As shown in FIG. For example, the processing unit may comprise a plurality of processors or one processor and one controller. Other processing configurations are also possible, such as a parallel processor.

The software may include a computer program, code, instructions, or a combination of one or more of the foregoing, and may be configured to configure the processing device to operate as desired or to process it collectively or collectively Device can be commanded. The software and / or data may be in the form of any type of machine, component, physical device, virtual equipment, computer storage media, or device , Or may be permanently or temporarily embodied in a transmitted signal wave. The software may be distributed over a networked computer system and stored or executed in a distributed manner. The software and data may be stored on one or more computer readable recording media.

The method according to an embodiment may be implemented in the form of a program command that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions to be recorded on the medium may be those specially designed and configured for the embodiments or may be available to those skilled in the art of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape; optical media such as CD-ROMs and DVDs; magnetic media such as floppy disks; Magneto-optical media, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments. For example, it is to be understood that the techniques described may be performed in a different order than the described methods, and / or that components of the described systems, structures, devices, circuits, Lt; / RTI &gt; or equivalents, even if it is replaced or replaced.

Therefore, other implementations, other embodiments, and equivalents to the claims are also within the scope of the following claims.

Claims (34)

Identifying relationship data relating to a relationship between a first object and a second object having an interpersonal relationship with the first object;
Generating an item-by-item record of a data field table for defining an interpersonal relationship type from the relationship data; And
Determining an interpersonal relationship type between the first object and the second object based on the record,
Wherein the step of determining the interpersonal relationship type comprises:
Defining a type of a statistic amount and a type determination criterion for determining the interpersonal relationship type;
Obtaining data necessary for calculating the statistic amount for each type of interpersonal relationship for which interpersonal relationship type is defined, each time the record is generated;
Calculating the statistic amount for records of a predetermined period to be analyzed; And
And comparing the calculated value of the statistic with the type determination criterion to determine an interpersonal relationship type
A method of providing coaching information through interpersonal relationship type interpersonal coaching system. The method according to claim 1,
The item of the data field table includes:
An object of the event, a synchronization of the event, information on a time when the event occurs, a content of the event, a result of the event, a space information of the event, , Demographic information on the relevant person of the event, status information of the relevant person of the event, current or future issue information inferred from the relationship data, information on the status of the first object or the second object Personal information &lt; RTI ID = 0.0 &gt;
A method of providing coaching information through interpersonal relationship type interpersonal coaching system. The method according to claim 1,
Determining the manner of providing coaching for the first object or the second object in consideration of the determined interpersonal relationship type
A method of providing coaching information through interpersonal relationship type interpersonal coaching system. The method of claim 3,
Wherein the coaching information includes information on a gift offering or behavior type recommendation that can be provided by the implementation of the first object to the second object
A method of providing coaching information through interpersonal relationship type interpersonal coaching system. Identifying relationship data relating to a relationship between a first object and a second object having an interpersonal relationship with the first object;
Generating an item-by-item record of a data field table for defining an interpersonal relationship type from the relationship data; And
Determining an interpersonal relationship type between the first object and the second object based on the record,
Wherein the step of determining the interpersonal relationship type comprises:
Calculating an index value for each interpersonal relationship type in which the interpersonal relationship type is defined, each time the record is generated;
Calculating an average of the index values for records of a predetermined period to be analyzed; And
Comparing the arithmetic mean of the exponent value and the exponential weighted average of the exponent value to calculate the trend of the interpersonal stage
A method of providing coaching information through interpersonal relationship type interpersonal coaching system. Identifying relationship data relating to a relationship between a first object and a second object having an interpersonal relationship with the first object;
Generating an item-by-item record of a data field table for defining an interpersonal relationship type from the relationship data; And
Determining an interpersonal relationship type between the first object and the second object based on the record,
Wherein the step of determining the interpersonal relationship type comprises calculating an index value for each interpersonal relationship type in which the interpersonal relationship type is defined each time the record is generated,
The calculation of the exponent value is performed by calculating the association index between the activity / activity occurrence time and the relationship type for each relationship type, calculating the association index between the activity / activity occurrence place and the relationship type for each relationship type, The process of calculating the association between the activity occurrence status and the relationship type, calculating the association index between the activity / activity subject and the related relationship type by the relationship type, and calculating the importance of each time, place, And a process of calculating an association index by applying the weighting factor
A method of providing coaching information through interpersonal relationship type interpersonal coaching system. Identifying relationship data relating to a relationship between a first object and a second object having an interpersonal relationship with the first object;
Generating an item-by-item record of a data field table for defining an interpersonal relationship type from the relationship data; And
Determining an interpersonal relationship type between the first object and the second object based on the record,
Wherein the step of determining the interpersonal relationship type comprises:
Calculating an index value for each interpersonal relationship type in which a plurality of interpersonal relationship types are defined mutually exclusive each time the record is generated;
Identifying an interpersonal relationship type having a sum of a maximum exponent value with a sum of the exponent values for a certain period to be analyzed; And
Determining an interpersonal type of the first object and the second object based on whether the sum of the maximum exponent values is greater than or equal to a predetermined threshold
A method of providing coaching information through interpersonal relationship type interpersonal coaching system. 8. The method of claim 7,
Wherein determining the interpersonal type of the first object and the second object based on whether the maximum exponent value is greater than or equal to a predetermined threshold comprises:
If the sum of the maximum exponent values is smaller than a predetermined threshold value, determining an interpersonal relationship type determination, and if the sum of the maximum exponent values is equal to or greater than the threshold value, And an interpersonal relationship type of the second object
A method of providing coaching information through interpersonal relationship type interpersonal coaching system. Identifying relationship data relating to a relationship between a first object and a second object having an interpersonal relationship with the first object;
Generating an item-by-item record of a data field table for defining an interpersonal relationship type from the relationship data; And
Determining an interpersonal relationship type between the first object and the second object based on the record,
Wherein the step of determining the interpersonal relationship type comprises:
Calculating an index percentage for each type of interpersonal relationship in which a plurality of interpersonal relationship types are not mutually exclusive;
Identifying an interpersonal relationship type having a maximum exponent percentage value with the largest exponent value for a certain period of time to be analyzed; And
Determining an interpersonal type of the first object and the second object based on whether the maximum exponent percentage value is greater than or equal to a predetermined threshold value
A method of providing coaching information through interpersonal relationship type interpersonal coaching system. 10. The method of claim 9,
Determining the interpersonal type of the first object and the second object based on whether the maximum exponent percentage value is greater than or equal to a predetermined threshold value,
Wherein if the maximum exponent percentage value is smaller than a predetermined threshold value, the interpersonal relationship type determination is suspended, and if the plurality of interpersonal relationship types having the exponent percentage value equal to or larger than the threshold value are plural, As the interpersonal type of the second object
A method of providing coaching information through interpersonal relationship type interpersonal coaching system. Identifying relationship data relating to a relationship between a first object and a second object having an interpersonal relationship with the first object;
Generating an item-by-item record of a data field table for defining an interpersonal relationship type from the relationship data; And
Determining an interpersonal relationship type between the first object and the second object based on the record,
Wherein the step of determining the interpersonal relationship type comprises:
Determining, whenever the record is newly generated, whether the newly generated record matches a plurality of interpersonal relationship types to record patterns derived by mutually exclusive interpersonal type;
Identifying an interpersonal relationship type having a maximum number of pattern counts having the largest number of patterns belonging to patterns of interpersonal relationship types, with respect to records generated for a predetermined period of time to be analyzed; And
Determining an interpersonal type of the first object and the second object based on whether the maximum pattern number value is greater than or equal to a preset threshold value
A method of providing coaching information through interpersonal relationship type interpersonal coaching system. 12. The method of claim 11,
Determining an interpersonal type of the first object and the second object based on whether the maximum pattern number value is greater than or equal to a preset threshold value,
Wherein if the maximum pattern number value is smaller than the predetermined threshold value, the interpersonal relationship type determination is suspended; and if the maximum pattern number value is greater than or equal to the threshold value, the interpersonal relationship type having the maximum pattern number value, 2 Determine by interpersonal type of object
A method of providing coaching information through interpersonal relationship type interpersonal coaching system. Identifying relationship data relating to a relationship between a first object and a second object having an interpersonal relationship with the first object;
Generating an item-by-item record of a data field table for defining an interpersonal relationship type from the relationship data; And
Determining an interpersonal relationship type between the first object and the second object based on the record,
Wherein the step of determining the interpersonal relationship type comprises:
Determining, for each new record created, whether the newly generated record matches a plurality of interpersonal relationship types that are derived for each interpersonal relationship type that is not mutually exclusive;
Identifying an interpersonal relationship type having a maximum pattern frequency percentage value with a maximum pattern frequency value for records generated for a predetermined period of time to be analyzed; And
Determining an interpersonal type of the first object and the second object based on whether the maximum pattern number percentage value is greater than or equal to a predetermined threshold value
A method of providing coaching information through interpersonal relationship type interpersonal coaching system. 14. The method of claim 13,
Determining the interpersonal type of the first object and the second object based on whether the maximum pattern number percentage value is equal to or greater than a predetermined threshold value,
Wherein if the percentage of the maximum number of patterns is less than a predetermined threshold value, the determination of the interpersonal relationship type is suspended, and if the plurality of interpersonal types having the maximum value of the pattern number percentages equal to or greater than the threshold value are plural, 1 &lt; / RTI &gt; object and the second object
A method of providing coaching information through interpersonal relationship type interpersonal coaching system. Identifying relationship data relating to a relationship between a first object and a second object having an interpersonal relationship with the first object;
Generating an item-by-item record of a data field table for defining an interpersonal relationship type from the relationship data; And
Determining an interpersonal relationship type between the first object and the second object based on the record,
Wherein the step of determining the interpersonal relationship type comprises:
The record is applied to one model that has been established to classify all interpersonal relationship types every time the record is generated so as to calculate an error rate for the corresponding interpersonal relationship types that are defined mutually exclusive of the plurality of interpersonal relationship types ;
Checking an interpersonal relationship type having a maximum error rate adjustment frequency value having a largest error rate adjusted frequency value based on the error rate for a predetermined period to be analyzed; And
Determining an interpersonal type of the first object and the second object based on whether the maximum error rate adjustment frequency value is greater than or equal to a preset threshold value
A method of providing coaching information through interpersonal relationship type interpersonal coaching system. 16. The method of claim 15,
Determining an interpersonal type of the first object and the second object based on whether the maximum error rate adjustment frequency value is equal to or greater than a preset threshold value,
Wherein if the maximum error rate adjustment frequency value is smaller than a predetermined threshold value, the interpersonal relationship type determination is suspended; and if the maximum error rate adjustment frequency value is greater than or equal to the threshold value, And an interpersonal relationship type of the second object
A method of providing coaching information through interpersonal relationship type interpersonal coaching system. Identifying relationship data relating to a relationship between a first object and a second object having an interpersonal relationship with the first object;
Generating an item-by-item record of a data field table for defining an interpersonal relationship type from the relationship data; And
Determining an interpersonal relationship type between the first object and the second object based on the record,
Wherein the step of determining the interpersonal relationship type comprises:
The record is applied to one model that has been established to classify all interpersonal relationship types every time the record is generated so that the error rate for the corresponding type of interpersonal relationship types in which plural interpersonal types are not defined mutually exclusive Calculating;
Identifying an interpersonal relationship type having a maximum frequency percentage value having a value of a frequency percentage based on the error rate for a certain period of time to be analyzed; And
Determining an interpersonal type of the first object and the second object based on whether the maximum frequency percentage value is greater than or equal to a predetermined threshold value
A method of providing coaching information through interpersonal relationship type interpersonal coaching system. 18. The method of claim 17,
Determining an interpersonal type of the first object and the second object based on whether the maximum frequency percentage value is greater than or equal to a preset threshold value,
If the maximum frequency percentage value is smaller than a predetermined threshold value, determining the interpersonal relationship type determination, and if the interpersonal relationship type having the frequency percentage value equal to or greater than the threshold value is a plurality of interpersonal relationship types, As the interpersonal type of the second object
A method of providing coaching information through interpersonal relationship type interpersonal coaching system. Identifying relationship data relating to a relationship between a first object and a second object having an interpersonal relationship with the first object;
Generating an item-by-item record of a data field table for defining an interpersonal relationship type from the relationship data; And
Determining an interpersonal relationship type between the first object and the second object based on the record,
Wherein the step of determining the interpersonal relationship type comprises:
For the similar similar object clusters, we identify the closest clusters of object 1 and object 2,
Calculating an error rate for the corresponding type of interpersonal relationship types in which a plurality of interpersonal relationship types are defined mutually exclusive by applying the record to the classification model established for the most similar community each time the record is generated ;
Checking an interpersonal relationship type having a maximum error rate adjustment frequency value having a largest error rate adjusted frequency value based on the error rate for a predetermined period to be analyzed; And
Determining an interpersonal type of the first object and the second object based on whether the maximum error rate adjustment frequency value is greater than or equal to a preset threshold value
A method of providing coaching information through interpersonal relationship type interpersonal coaching system. 20. The method of claim 19,
Determining an interpersonal type of the first object and the second object based on whether the maximum error rate adjustment frequency value is equal to or greater than a preset threshold value,
Wherein if the maximum error rate adjustment frequency value is smaller than a predetermined threshold value, the interpersonal relationship type determination is suspended; and if the maximum error rate adjustment frequency value is greater than or equal to the threshold value, And an interpersonal relationship type of the second object
A method of providing coaching information through interpersonal relationship type interpersonal coaching system. Identifying relationship data relating to a relationship between a first object and a second object having an interpersonal relationship with the first object;
Generating an item-by-item record of a data field table for defining an interpersonal relationship type from the relationship data; And
Determining an interpersonal relationship type between the first object and the second object based on the record,
Wherein the step of determining the interpersonal relationship type comprises:
For the similar similar object clusters, we identify the closest clusters of object 1 and object 2,
Each time the record is generated, the record is applied to the classification model established for the most similar clusters to calculate the error rate for the corresponding type among the interpersonal types for which a plurality of interpersonal types are not defined mutually exclusive step;
Identifying an interpersonal relationship type having a maximum frequency percentage value having a value of a frequency percentage based on the error rate for a certain period of time to be analyzed; And
Determining an interpersonal type of the first object and the second object based on whether the maximum frequency percentage value is greater than or equal to a predetermined threshold value
A method of providing coaching information through interpersonal relationship type interpersonal coaching system. 22. The method of claim 21,
Determining an interpersonal type of the first object and the second object based on whether the maximum frequency percentage value is greater than or equal to a preset threshold value,
If the maximum frequency percentage value is smaller than a predetermined threshold value, determining the interpersonal relationship type determination, and if the interpersonal relationship type having the frequency percentage value equal to or greater than the threshold value is a plurality of interpersonal relationship types, As the interpersonal type of the second object
A method of providing coaching information through interpersonal relationship type interpersonal coaching system.
Establish as many models as interpersonal type by community Identifying relationship data relating to a relationship between a first object and a second object having an interpersonal relationship with the first object;
Generating an item-by-item record of a data field table for defining an interpersonal relationship type from the relationship data; And
Determining an interpersonal relationship type between the first object and the second object based on the record,
Wherein the step of determining the interpersonal relationship type comprises:
For similar object clusters by predefined interpersonal type, we identify the closest clusters of object 1 and object 2, one by interpersonal type,
Wherein each time the record is generated, the record is applied to each of the classification models for the corresponding interpersonal type established for each of the most similar clusters, so that a plurality of interpersonal relationship types are classified into mutually exclusively defined interpersonal types, rate;
Identifying an interpersonal relationship type having a value of a minimum average error rate with a minimum average of the error rates for a predetermined period of time to be analyzed; And
Determining an interpersonal type of the first object and the second object based on whether the minimum average error rate value is less than or equal to a predetermined threshold value
A method of providing coaching information through interpersonal relationship type interpersonal coaching system. 24. The method of claim 23,
Determining an interpersonal type of the first object and the second object based on whether the minimum average error rate value is less than or equal to a predetermined threshold value,
Wherein if the minimum average error rate value is greater than a preset threshold value, the interpersonal relationship type determination is suspended; and if the minimum average error rate value is less than or equal to the threshold value, the interpersonal relationship type having the minimum average error rate value, Determine by interpersonal type of object
A method of providing coaching information through interpersonal relationship type interpersonal coaching system. Identifying relationship data relating to a relationship between a first object and a second object having an interpersonal relationship with the first object;
Generating an item-by-item record of a data field table for defining an interpersonal relationship type from the relationship data; And
Determining an interpersonal relationship type between the first object and the second object based on the record,
Wherein the step of determining the interpersonal relationship type comprises:
For similar object clusters by predefined interpersonal type, we identify the closest clusters of object 1 and object 2, one by interpersonal type,
Each time the record is generated, the record is applied to each of the classification models for each corresponding interpersonal type established for each of the most similar clusters, thereby calculating an error rate for each type of interpersonal relationship in which a plurality of interpersonal types are not defined mutually exclusive ;
Confirming an interpersonal relationship type having a minimum error rate percentage value having a smallest value of an error rate percentage based on the error rate for a predetermined period to be analyzed; And
Determining an interpersonal type of the first object and the second object based on whether the minimum error rate percent value is less than or equal to a predetermined threshold value
A method of providing coaching information through interpersonal relationship type interpersonal coaching system. 26. The method of claim 25,
Determining the interpersonal type of the first object and the second object based on whether the minimum error rate percent value is less than or equal to a predetermined threshold value,
Wherein if the minimum error rate percent value is greater than a predetermined threshold value, the determination of the interpersonal relationship type is suspended, and if the plurality of interpersonal relationship types having the error rate percentage value equal to or less than the threshold value are plural, The type of interpersonal relationship of the second object
A method of providing coaching information through interpersonal relationship type interpersonal coaching system. The method according to claim 1,
Determining a method of providing coaching for the first object in consideration of the determined interpersonal relationship type
A method of providing coaching information through interpersonal relationship type interpersonal coaching system. 28. The method of claim 27,
The method of claim 1,
A first procedure for inferring a relationship forming step between the first object and the second object using the relational data, a second procedure for detecting a situation in which coaching application of the second object is required, A third procedure and a fourth procedure for grasping an acceptable level of coaching of the first object
A method of providing coaching information through interpersonal relationship type interpersonal coaching system. 29. The method of claim 28,
The fourth procedure includes the steps of identifying behavior history data of the first object and the second object included in the relational data and subdividing the behavior history by the relation type from the behavior history data, The process of classifying behavior history by process, relation trend, the process of extracting behavior type pattern for each combination of relationship type, relation formation step and relation trend, A process of extracting a behavior providing path pattern for each combination of a relationship type, a relationship formation step and a relationship trend, or a process of providing a behavior providing method for each relationship type / Including a course
A method of providing coaching information through interpersonal relationship type interpersonal coaching system. An interpersonal type database in which an interpersonal relationship type is defined;
A data field table for defining the interpersonal relationship type;
A relational data obtaining unit for obtaining relational data related to a relationship between a first object and a second object related to the first object;
A record generating unit for generating a record for each item of the data field table from the relational data; And
And an interpersonal relationship type determination unit for determining an interpersonal relationship type between the first object and the second object based on the record,
The interpersonal relationship type determination unit
The data type and the type determination criterion for determining the interpersonal relationship type are defined and data necessary for calculating the statistic amount for each interpersonal relationship type in which the interpersonal relationship type is defined is acquired every time the record is generated, The statistical amount is calculated for the records of a certain period of time to be a target, and the interpersonal relationship type is determined by comparing the calculated value of the statistic with the type determination criterion
Coaching information providing device through interpersonal type identification. 31. The method of claim 30,
And a coaching method determining unit determining a method of providing coaching for the first object or the second object in consideration of the determined interpersonal relationship type
Coaching information providing device through interpersonal type identification. An interpersonal type database in which an interpersonal relationship type is defined;
A data field table for defining the interpersonal relationship type;
A relational data obtaining unit for obtaining relational data related to a relationship between a first object and a second object related to the first object;
A record generating unit for generating a record for each item of the data field table from the relational data; And
And an interpersonal relationship type determination unit for determining an interpersonal relationship type between the first object and the second object based on the record,
Wherein the interpersonal relationship type determining unit determines,
Calculating an index value for each type of interpersonal relationship each time the record is generated, calculating an average of the index values for records of a predetermined period to be analyzed, The exponential weighted average of the index values is compared to calculate the trend of the interpersonal step
Coaching information providing device through interpersonal type identification. An interpersonal type database in which an interpersonal relationship type is defined;
A data field table for defining the interpersonal relationship type;
A relational data obtaining unit for obtaining relational data related to a relationship between a first object and a second object related to the first object;
A record generating unit for generating a record for each item of the data field table from the relational data; And
And an interpersonal relationship type determination unit for determining an interpersonal relationship type between the first object and the second object based on the record,
Wherein the interpersonal relationship type determining unit determines,
Each time the record is newly generated, it is determined whether or not the newly generated record is matched to the record patterns derived for each interpersonal type, and for each record generated for a predetermined period of time to be analyzed, Identifying an interpersonal relationship type having a maximum number of pattern times having the largest number of times belonging to the patterns of each relationship type and determining an interpersonal relationship type To determine the type
Coaching information providing device through interpersonal type identification. An interpersonal type database in which an interpersonal relationship type is defined;
A data field table for defining the interpersonal relationship type;
A relational data obtaining unit for obtaining relational data related to a relationship between a first object and a second object related to the first object;
A record generating unit for generating a record for each item of the data field table from the relational data; And
And an interpersonal relationship type determination unit for determining an interpersonal relationship type between the first object and the second object based on the record,
Wherein the interpersonal relationship type determining unit determines,
Each time the record is generated, the record is applied to the pre-established interpersonal relationship classification model to determine the interpersonal type of the first object and the second object based on the error rate of the classification model
Coaching information providing device through interpersonal type identification.

Images