CN111612581A - Method, device and equipment for recommending articles and storage medium - Google Patents
Method, device and equipment for recommending articles and storage medium Download PDFInfo
- Publication number
- CN111612581A CN111612581A CN202010419213.3A CN202010419213A CN111612581A CN 111612581 A CN111612581 A CN 111612581A CN 202010419213 A CN202010419213 A CN 202010419213A CN 111612581 A CN111612581 A CN 111612581A
- Authority
- CN
- China
- Prior art keywords
- recommended
- item
- score
- determining
- article
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000012937 correction Methods 0.000 claims abstract description 56
- 238000003062 neural network model Methods 0.000 claims abstract description 15
- 238000004590 computer program Methods 0.000 claims description 6
- 230000003287 optical effect Effects 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 238000012986 modification Methods 0.000 description 5
- 230000004048 modification Effects 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 239000002537 cosmetic Substances 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 239000013307 optical fiber Substances 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention discloses a method, a device, equipment and a storage medium for recommending articles. Wherein, the method comprises the following steps: inputting user information and attribute information of an article to be recommended into a pre-trained deep neural network model, and determining a model score of the article to be recommended; determining a correction score of the item to be recommended according to the heat type of the item to be recommended and/or the attribute information of the item to be recommended; and selecting a target item for the user from the items to be recommended according to the model score and the correction score of the items to be recommended. According to the embodiment of the invention, the target object recommended for the user is determined by calculating the model score and the correction score, so that the accuracy and the flexibility of object recommendation are improved, targeted recommendation to the user is realized, and the user experience is improved.
Description
Technical Field
The embodiment of the invention relates to computer technology, in particular to a method, a device, equipment and a storage medium for recommending articles.
Background
With the popularization of online shopping, the demand of users for item recommendation is higher and higher, and the targeted item recommendation for the users can meet the purchase demand of the users and promote item consumption.
In the existing recommendation system, all items related to a user are recalled, then the relevance between the user and the items is calculated, and ranking is performed according to the calculated scores, so as to select the items for recommendation.
However, the method cannot perform priority recommendation for dimensions such as sales volume and click volume of the articles, for example, the articles clicked in real time cannot be prioritized in front, the recommendation flexibility is not sufficient, and the recommended articles cannot accurately meet the user requirements.
Disclosure of Invention
The embodiment of the invention provides a method, a device, equipment and a storage medium for recommending articles, so as to realize accurate article recommendation for a user and improve user experience.
In a first aspect, an embodiment of the present invention provides an item recommendation method, where the method includes:
inputting user information and attribute information of an article to be recommended into a pre-trained deep neural network model, and determining a model score of the article to be recommended;
determining a correction score of the item to be recommended according to the heat type of the item to be recommended and/or the attribute information of the item to be recommended;
and selecting a target item for the user from the items to be recommended according to the model score and the correction score of the items to be recommended.
In a second aspect, an embodiment of the present invention further provides an article recommendation apparatus, where the apparatus includes:
the model score determining module is used for inputting the user information and the attribute information of the to-be-recommended article into a pre-trained deep neural network model and determining the model score of the to-be-recommended article;
the correction score determining module is used for determining the correction score of the item to be recommended according to the heat degree type of the item to be recommended and/or the attribute information of the item to be recommended;
and the target article selection module is used for selecting a target article for the user from the articles to be recommended according to the model score and the correction score of the articles to be recommended.
In a third aspect, an embodiment of the present invention further provides a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the processor implements the item recommendation method according to any embodiment of the present invention.
In a fourth aspect, embodiments of the present invention further provide a storage medium containing computer-executable instructions, which when executed by a computer processor, are configured to perform the item recommendation method according to any of the embodiments of the present invention.
According to the embodiment of the invention, the model score of the article is calculated through a pre-trained deep neural network model, the correction score of the article is determined according to the heat type and/or attribute information of the article, and the article finally recommended to the user is selected according to the model score and the correction score. The problem of among the prior art, can only calculate the score of article through single dimension, article recommendation inaccurate that leads to is solved, through calculating the correction score, improved article recommendation's accuracy nature and flexibility, realize the recommendation to the user pertinence, promote user experience.
Drawings
Fig. 1 is a schematic flow chart of an item recommendation method according to a first embodiment of the present invention;
FIG. 2 is a flowchart illustrating an item recommendation method according to a second embodiment of the present invention;
fig. 3 is a block diagram of an article recommendation device according to a third embodiment of the present invention;
fig. 4 is a schematic structural diagram of a computer device in the fourth embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.
Example one
Fig. 1 is a flowchart illustrating an article recommendation method according to an embodiment of the present invention, where the embodiment is applicable to automatically recommending an article for a user, and the method may be executed by an article recommendation apparatus. As shown in fig. 1, the method specifically includes the following steps:
The user information may include information such as gender, occupation, age and the like of the user, and the attribute information of the item to be recommended may include information such as a score, inventory, click rate and volume of the item to be recommended. The pre-trained deep neural network model can be used for calculating the recommended probability of the to-be-recommended articles, the user information and the attribute information of the to-be-recommended articles are input data of the model, the model score is output data of the model, and the model score can represent the recommended probability of the to-be-recommended articles. And calculating a model score of the item to be recommended according to the user information and the attribute information of the item to be recommended, wherein the higher the model score is, the higher the possibility that the item to be recommended is. For example, if the item to be recommended is a cosmetic, the model score obtained when the gender of the user is female is higher than the model score obtained when the gender of the user is male, that is, the cosmetic is more likely to be recommended to the female user than to the male user.
And step 120, determining a correction score of the item to be recommended according to the heat type of the item to be recommended and/or the attribute information of the item to be recommended.
The popularity type can be determined by the click amount and the click time of the to-be-recommended articles, and can be divided into real-time clicking, real-time hot selling, historical clicking, bottom-pocketing hot selling and the like. The real-time click type items are items clicked within 24 hours by the user, the real-time hot-sell type items are similar and complementary items of the items clicked within 24 hours by the user, the historical click type items are items clicked beyond 24 hours by the user, and the bottom-pocketed hot-sell type items are full-web hot-sell items.
The correction score is used for correcting the model score, and the correction score of at least one item to be recommended can be determined according to the heat type of the item to be recommended and/or the attribute information of the item to be recommended. The popularity type and/or the attribute information of the to-be-recommended article can be associated with the preset correction score, after the to-be-recommended article is obtained, the popularity type and/or the attribute information of the to-be-recommended article can be determined, and the correction score associated with the popularity type and/or the attribute information of the to-be-recommended article is searched. Correction scores of different hot type items may be preset, for example, the correction score of the real-time click type item is set to 10 points, the correction score of the real-time hot-sell type item is set to 7 points, the correction score of the historical click type item is set to 4 points, and the correction score of the bottom-pocketed hot-sell type item is set to 1 point. The modification score may also be preset according to the attribute information of the item to be recommended, for example, if the click rate of the item to be recommended exceeds 1 ten thousand, the modification score is set to 5, and if the volume of the item to be recommended exceeds 2 ten thousand, the modification score is set to 10.
And step 130, selecting a target item for the user from the items to be recommended according to the model score and the correction score of the items to be recommended.
The modification score may be added or subtracted to the model score, for example, when the click rate of the item to be recommended is less than 5000, the modification score may be set to-3, and if the model score of the item to be recommended is 7, the final score of the item to be recommended is 4. If the correction score obtained according to the heat type of the object to be recommended is 5 scores and the correction score obtained according to the attribute information of the object to be recommended is-2 scores, the two correction scores can be subjected to weighted summation to obtain a final correction score, and then the final correction score of the object to be recommended is obtained by calculating the final correction score and the model score.
The model score can be added or subtracted according to the correction score, or the weights of the correction score and the model score can be set firstly, and then the final score of the item to be recommended is calculated.
And the target item is the item finally recommended to the user, and whether the item to be recommended is selected as the target item to be recommended can be determined according to the final score. For example, a score threshold of the final score may be preset to be 7, and if the final score exceeds the score threshold, it is determined that the item to be recommended is the target item.
According to the technical scheme of the embodiment, the model score of the article is calculated through a pre-trained deep neural network model, the correction score of the article is determined according to the heat type and/or attribute information of the article, and the article finally recommended to the user is selected according to the model score and the correction score. The problem of among the prior art, can only calculate the score of article through single dimension, article recommendation inaccurate that leads to is solved, through calculating the correction score, improved article recommendation's accuracy nature and flexibility, realize the recommendation to the user pertinence, promote user experience.
Example two
Fig. 2 is a flowchart illustrating an item recommendation method according to a second embodiment of the present invention, which is further optimized based on the second embodiment. As shown in fig. 2, the method specifically includes the following steps:
And step 220, determining the recall score of the item to be recommended according to the popularity type of the item to be recommended.
Wherein the revised score may be categorized as a recall score and a press score. The recall score is a modified score obtained according to the popularity of the item to be recommended, different popularity types can be associated with the recall score, and the higher the popularity of the item to be recommended is, the higher the recall score can be set. For example, the recall score of the real-time click type item is set to 10, the recall score of the real-time hot-sell type item is set to 7, the recall score of the historical click type item is set to 4, and the recall score of the bottom hot-sell type item is set to 1.
In this embodiment, optionally, determining a recall score of the item to be recommended according to the popularity type of the item to be recommended includes: determining a target recall data table to which the candidate item data table belongs according to the heat type of the item to be recommended; the popularity type is determined by the click quantity and the click time of the item to be recommended; and determining the recall score of the item to be recommended according to the target recall data table.
Specifically, the candidate item data table is an acquired data table containing items to be recommended, and the item set in the target recall data table is a subset of the item set in the candidate item data table. The target recall data table may be divided according to a popularity type, the popularity type is determined by the click amount and the click time of the item to be recommended, for example, the popularity type is divided into a real-time click, a real-time hot-sell, a historical click and a bottom-of-the-pocket hot-sell, and the target recall data table is divided into a real-time click data table, a real-time hot-sell data table, a historical click data table and a bottom-of-the-pocket hot-sell data table. The recall score of each target recall data table is set in advance, for example, the recall score of the real-time click data table is set to 10. When the to-be-recommended item is acquired, determining the popularity type of the to-be-recommended item, for example, if the popularity type of the to-be-recommended item belongs to real-time clicking, adding the to-be-recommended item to a real-time clicking data table, wherein the recall scores of the items in the real-time clicking data table are all 10 scores, and therefore the recall score of the to-be-recommended item is 10 scores.
By dividing the target recall data table and adding the to-be-recommended articles into the corresponding target recall data table, the recall scores of the to-be-recommended articles can be quickly obtained, the article recommendation efficiency is effectively improved, the article recommendation accuracy can be improved by calculating the recall scores, and the user experience is improved.
And step 230, determining the pressing score of the item to be recommended according to the attribute information of the item to be recommended.
The suppressing score is a correction score obtained according to the attribute information of the item to be recommended, the attribute information of the item to be recommended can include stock and score, and the suppressing score can be divided into the stock suppressing score and the score suppressing score according to the attribute information of the item to be recommended. The pressing score can be set to be a negative number, the pressing score is associated with the inventory and the score of the item to be recommended, and the fewer the inventory and/or the lower the score of the item to be recommended, the higher the pressing score and the lower the final score of the item to be recommended.
In this embodiment, optionally, determining the pressing score of the item to be recommended according to the attribute information of the item to be recommended includes: and if the to-be-recommended article is determined to belong to the inventory shortage data table according to the attribute information of the to-be-recommended article, determining the inventory pressing score of the to-be-recommended article according to the inventory quantity of the to-be-recommended article.
Specifically, at least one inventory shortage data table is pre-established, after the item to be recommended is obtained, the inventory of the item to be recommended is determined, and if the inventory is smaller than an inventory threshold value, the item to be recommended is added to the inventory shortage data table corresponding to the inventory threshold value. The method includes the steps that inventory pressing scores of articles in an inventory shortage data table are preset, for example, an inventory threshold value is set to be 300, the inventory pressing score corresponding to the inventory threshold value is-3, the inventory threshold value is set to be 500, the inventory pressing score corresponding to the inventory threshold value is-1, the obtained inventory of the articles to be recommended is 100, the articles to be recommended are divided into the inventory shortage data table with the inventory threshold value of 300, and the inventory pressing score of the articles to be recommended is-3 according to the preset inventory pressing score. If the inventory of the item to be recommended exceeds an inventory threshold, the inventory tie-down score may not be calculated. By establishing the inventory shortage database and setting the inventory suppressing score, the suppressing score of the article to be recommended can be quickly obtained, the calculation process is reduced, the article is recommended from the inventory perspective, and the article recommendation efficiency and the article recommendation precision can be improved.
In this embodiment, optionally, determining the pressing score of the item to be recommended according to the attribute information of the item to be recommended includes: and if the to-be-recommended article is determined to belong to the blacklist data table according to the attribute information of the to-be-recommended article, determining the score of the to-be-recommended article according to the score of the to-be-recommended article.
Specifically, at least one blacklist data table is pre-established, after the item to be recommended is obtained, the score of the item to be recommended is determined, and if the score is smaller than a score threshold value, the item to be recommended is added to the blacklist data table corresponding to the score threshold value. The method includes the steps that a grading and pressing score of an article in a blacklist data table is preset, for example, a grading threshold value is set to be 3 points, the grading and pressing score is-3 points, the obtained grade of the article to be recommended is 2 points, the article to be recommended is divided into the blacklist data table, and the inventory pressing score of the article to be recommended is-3 points according to the preset grading and pressing score. If the score of the item to be recommended exceeds the score threshold, the score suppressing score may not be calculated. By establishing the blacklist database and setting the grading and suppressing score, the suppressing score of the object to be recommended can be quickly obtained, the calculation process is reduced, the object is recommended from the grading angle, and the object recommendation efficiency and the object recommendation precision can be improved.
At least one inventory shortage data table and at least one blacklist data table can be preset, different inventory suppressing scores can be set for the inventory shortage data tables with different inventory quantity grades, and different grading suppressing scores can be set for the blacklist data tables with different grading grades. The order of determination of the stock suppressing score and the score suppressing score is not limited.
And step 240, determining a correction score of the item to be recommended according to the recall score and the pressing score of the item to be recommended.
And if the correction score of the item to be recommended only has a recall score or a suppressing score, performing integrated calculation on the correction score and the model score according to a preset weight or other calculation methods to obtain a final score. And if the revised score of the recommended item comprises a recall score and a pressing score, integrating the recall score and the pressing score into a revised score, and then performing integrated calculation with the model score. For example, the recall score and the stroke score may be weighted and summed to provide a revised score.
And step 250, selecting a target article for the user from the articles to be recommended according to the model score and the correction score of the articles to be recommended.
According to the embodiment of the invention, the model score of an article is calculated through a pre-trained deep neural network model, the recall score is determined according to the popularity type of the article, the suppressing score is determined according to the attribute information, the correcting score of the article is determined by the recall score and the suppressing score, and the article finally recommended to a user is selected according to the model score and the correcting score. The problem of among the prior art, can only calculate the score of article through single dimension, article recommendation inaccurate that leads to is solved, through calculating the revised score of multidimension degree, improved article recommendation's accuracy and flexibility, realize the targeted recommendation to the user, promote user experience.
EXAMPLE III
Fig. 3 is a block diagram of an article recommendation apparatus according to a third embodiment of the present invention, which is capable of executing an article recommendation method according to any embodiment of the present invention, and includes functional modules and beneficial effects corresponding to the execution method. As shown in fig. 3, the apparatus specifically includes:
the model score determining module 301 is configured to input the user information and the attribute information of the to-be-recommended item into a pre-trained deep neural network model, and determine a model score of the to-be-recommended item;
a modified score determining module 302, configured to determine a modified score of the item to be recommended according to the heat type of the item to be recommended and/or the attribute information of the item to be recommended;
and the target item selection module 303 is configured to select a target item for the user from the items to be recommended according to the model score and the correction score of the items to be recommended.
Optionally, the modified score determining module 302 includes:
the recall score determining unit is used for determining the recall score of the item to be recommended according to the popularity type of the item to be recommended;
the pressing score determining unit is used for determining the pressing score of the item to be recommended according to the attribute information of the item to be recommended;
and the correction score obtaining unit is used for determining the correction score of the item to be recommended according to the recall score and the pressing score of the item to be recommended.
Optionally, the recall score determining unit is specifically configured to:
determining a target recall data table to which the candidate item data table belongs according to the heat type of the item to be recommended; wherein the heat type is determined by the click quantity and the click time of the item to be recommended;
and determining the recall score of the item to be recommended according to the target recall data table.
Optionally, the pressing fraction determining unit includes:
and the inventory pressing score determining unit is used for determining the inventory pressing score of the item to be recommended according to the inventory quantity of the item to be recommended if the item to be recommended is determined to belong to the inventory shortage data table according to the attribute information of the item to be recommended.
Optionally, the pressing fraction determining unit further includes:
and the score pressing and scoring determining unit is used for determining the score pressing and scoring of the to-be-recommended article according to the score of the to-be-recommended article if the to-be-recommended article is determined to belong to the blacklist data table according to the attribute information of the to-be-recommended article.
According to the embodiment of the invention, the model score of the article is calculated through a pre-trained deep neural network model, the correction score of the article is determined according to the heat type and/or attribute information of the article, and the article finally recommended to the user is selected according to the model score and the correction score. The problem of among the prior art, can only calculate the score of article through single dimension, article recommendation inaccurate that leads to is solved, through calculating the correction score, improved article recommendation's accuracy nature and flexibility, realize the recommendation to the user pertinence, promote user experience.
Example four
Fig. 4 is a schematic structural diagram of a computer device according to a fourth embodiment of the present invention. FIG. 4 illustrates a block diagram of an exemplary computer device 400 suitable for use in implementing embodiments of the present invention. The computer device 400 shown in fig. 4 is only an example and should not bring any limitations to the functionality or scope of use of the embodiments of the present invention.
As shown in fig. 4, computer device 400 is in the form of a general purpose computing device. The components of computer device 400 may include, but are not limited to: one or more processors or processing units 401, a system memory 402, and a bus 403 that couples the various system components (including the system memory 402 and the processing unit 401).
The system memory 402 may include computer system readable media in the form of volatile memory, such as Random Access Memory (RAM)404 and/or cache memory 405. The computer device 400 may further include other removable/non-removable, volatile/nonvolatile computer system storage media. By way of example only, storage system 406 may be used to read from and write to non-removable, nonvolatile magnetic media (not shown in FIG. 4, and commonly referred to as a "hard drive"). Although not shown in FIG. 4, a magnetic disk drive for reading from and writing to a removable, nonvolatile magnetic disk (e.g., a "floppy disk") and an optical disk drive for reading from or writing to a removable, nonvolatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 403 by one or more data media interfaces. Memory 402 may include at least one program product having a set (e.g., at least one) of program modules that are configured to carry out the functions of embodiments of the invention.
A program/utility 408 having a set (at least one) of program modules 407 may be stored, for example, in memory 402, such program modules 407 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which examples or some combination thereof may comprise an implementation of a network environment. Program modules 407 generally perform the functions and/or methods of the described embodiments of the invention.
The computer device 400 may also communicate with one or more external devices 409 (e.g., keyboard, pointing device, display 410, etc.), with one or more devices that enable a user to interact with the computer device 400, and/or with any devices (e.g., network card, modem, etc.) that enable the computer device 400 to communicate with one or more other computing devices. Such communication may be through input/output (I/O) interface 411. Moreover, computer device 400 may also communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network such as the Internet) via network adapter 412. As shown, network adapter 412 communicates with the other modules of computer device 400 over bus 403. It should be appreciated that although not shown in the figures, other hardware and/or software modules may be used in conjunction with computer device 400, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, among others.
The processing unit 401 executes various functional applications and data processing by executing programs stored in the system memory 402, for example, to implement an item recommendation method provided by an embodiment of the present invention, including:
inputting user information and attribute information of an article to be recommended into a pre-trained deep neural network model, and determining a model score of the article to be recommended;
determining a correction score of the item to be recommended according to the heat type of the item to be recommended and/or the attribute information of the item to be recommended;
and selecting a target item for the user from the items to be recommended according to the model score and the correction score of the items to be recommended.
EXAMPLE five
The fifth embodiment of the present invention further provides a storage medium containing computer-executable instructions, where the storage medium stores a computer program, and when the computer program is executed by a processor, the method for recommending an article according to the fifth embodiment of the present invention is implemented, where the method includes:
inputting user information and attribute information of an article to be recommended into a pre-trained deep neural network model, and determining a model score of the article to be recommended;
determining a correction score of the item to be recommended according to the heat type of the item to be recommended and/or the attribute information of the item to be recommended;
and selecting a target item for the user from the items to be recommended according to the model score and the correction score of the items to be recommended.
Computer storage media for embodiments of the invention may employ any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
A computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.
Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
Computer program code for carrying out operations for aspects of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (10)
1. An item recommendation method, comprising:
inputting user information and attribute information of an article to be recommended into a pre-trained deep neural network model, and determining a model score of the article to be recommended;
determining a correction score of the item to be recommended according to the heat type of the item to be recommended and/or the attribute information of the item to be recommended;
and selecting a target item for the user from the items to be recommended according to the model score and the correction score of the items to be recommended.
2. The method according to claim 1, wherein determining the revised score of the item to be recommended according to the heat type of the item to be recommended and the attribute information of the item to be recommended comprises:
determining a recall score of the item to be recommended according to the popularity type of the item to be recommended;
determining the pressing score of the item to be recommended according to the attribute information of the item to be recommended;
and determining a correction score of the item to be recommended according to the recall score and the suppressing score of the item to be recommended.
3. The method of claim 2, wherein determining the recall score of the item to be recommended according to the popularity type of the item to be recommended comprises:
determining a target recall data table to which the candidate item data table belongs according to the heat type of the item to be recommended; wherein the heat type is determined by the click quantity and the click time of the item to be recommended;
and determining the recall score of the item to be recommended according to the target recall data table.
4. The method according to claim 2, wherein determining the pressing score of the item to be recommended according to the attribute information of the item to be recommended comprises:
and if the item to be recommended is determined to belong to the inventory shortage data table according to the attribute information of the item to be recommended, determining the inventory pressing score of the item to be recommended according to the inventory quantity of the item to be recommended.
5. The method according to claim 2, wherein determining the pressing score of the item to be recommended according to the attribute information of the item to be recommended comprises:
and if the item to be recommended is determined to belong to a blacklist data table according to the attribute information of the item to be recommended, determining the score of the item to be recommended according to the score of the item to be recommended and the suppressing score.
6. An item recommendation device, comprising:
the model score determining module is used for inputting the user information and the attribute information of the to-be-recommended article into a pre-trained deep neural network model and determining the model score of the to-be-recommended article;
the correction score determining module is used for determining the correction score of the item to be recommended according to the heat degree type of the item to be recommended and/or the attribute information of the item to be recommended;
and the target article selection module is used for selecting a target article for the user from the articles to be recommended according to the model score and the correction score of the articles to be recommended.
7. The apparatus of claim 6, wherein the revision score determination module comprises:
the recall score determining unit is used for determining the recall score of the item to be recommended according to the heat degree type of the item to be recommended;
the pressing score determining unit is used for determining the pressing score of the to-be-recommended article according to the attribute information of the to-be-recommended article;
and the correction score obtaining unit is used for determining the correction score of the to-be-recommended article according to the recall score and the suppressing score of the to-be-recommended article.
8. The apparatus of claim 7, wherein the recall score determination unit is specifically configured to:
determining a target recall data table to which the candidate item data table belongs according to the heat type of the item to be recommended; wherein the heat type is determined by the click quantity and the click time of the item to be recommended;
and determining the recall score of the item to be recommended according to the target recall data table.
9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the item recommendation method as claimed in any one of claims 1-5 when executing the program.
10. A storage medium containing computer-executable instructions for performing the item recommendation method of any one of claims 1-5 when executed by a computer processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010419213.3A CN111612581A (en) | 2020-05-18 | 2020-05-18 | Method, device and equipment for recommending articles and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010419213.3A CN111612581A (en) | 2020-05-18 | 2020-05-18 | Method, device and equipment for recommending articles and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111612581A true CN111612581A (en) | 2020-09-01 |
Family
ID=72198354
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010419213.3A Pending CN111612581A (en) | 2020-05-18 | 2020-05-18 | Method, device and equipment for recommending articles and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111612581A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112257168A (en) * | 2020-10-30 | 2021-01-22 | 贝壳技术有限公司 | Recommendation method and device for single-space placement item set and computer-readable storage medium |
| CN113744016A (en) * | 2020-11-04 | 2021-12-03 | 北京沃东天骏信息技术有限公司 | Object recommendation method and device, equipment and storage medium |
| CN113763108A (en) * | 2021-01-29 | 2021-12-07 | 北京京东拓先科技有限公司 | Article pushing method and device, electronic equipment and storage medium |
| CN113763089A (en) * | 2020-10-27 | 2021-12-07 | 北京沃东天骏信息技术有限公司 | Article recommendation method and device and computer-readable storage medium |
| CN113781139A (en) * | 2020-10-19 | 2021-12-10 | 北京沃东天骏信息技术有限公司 | Item recommendation method, item recommendation device, equipment and medium |
| CN114022248A (en) * | 2021-11-08 | 2022-02-08 | 中国建设银行股份有限公司 | Product recommendation calculation method, device and equipment |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106454536A (en) * | 2016-09-19 | 2017-02-22 | 广州视源电子科技股份有限公司 | Method and device for determining information recommendation degree |
| CN106708938A (en) * | 2016-11-18 | 2017-05-24 | 北京大米科技有限公司 | Method and device for assisting recommendation |
| CN108334575A (en) * | 2018-01-23 | 2018-07-27 | 北京三快在线科技有限公司 | A kind of recommendation results sequence modification method and device, electronic equipment |
| CN109146193A (en) * | 2018-09-05 | 2019-01-04 | 平安科技（深圳）有限公司 | Product intelligent recommended method, device, computer equipment and storage medium |
| CN109903103A (en) * | 2017-12-07 | 2019-06-18 | 华为技术有限公司 | A kind of method and apparatus for recommending article |
| CN110147497A (en) * | 2019-05-15 | 2019-08-20 | 中国搜索信息科技股份有限公司 | A kind of personalization content recommendation method towards younger population |
| CN110458637A (en) * | 2019-06-19 | 2019-11-15 | 中国平安财产保险股份有限公司 | Product method for pushing and its relevant device neural network based |
-
2020
- 2020-05-18 CN CN202010419213.3A patent/CN111612581A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106454536A (en) * | 2016-09-19 | 2017-02-22 | 广州视源电子科技股份有限公司 | Method and device for determining information recommendation degree |
| CN106708938A (en) * | 2016-11-18 | 2017-05-24 | 北京大米科技有限公司 | Method and device for assisting recommendation |
| CN109903103A (en) * | 2017-12-07 | 2019-06-18 | 华为技术有限公司 | A kind of method and apparatus for recommending article |
| CN108334575A (en) * | 2018-01-23 | 2018-07-27 | 北京三快在线科技有限公司 | A kind of recommendation results sequence modification method and device, electronic equipment |
| CN109146193A (en) * | 2018-09-05 | 2019-01-04 | 平安科技（深圳）有限公司 | Product intelligent recommended method, device, computer equipment and storage medium |
| CN110147497A (en) * | 2019-05-15 | 2019-08-20 | 中国搜索信息科技股份有限公司 | A kind of personalization content recommendation method towards younger population |
| CN110458637A (en) * | 2019-06-19 | 2019-11-15 | 中国平安财产保险股份有限公司 | Product method for pushing and its relevant device neural network based |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113781139A (en) * | 2020-10-19 | 2021-12-10 | 北京沃东天骏信息技术有限公司 | Item recommendation method, item recommendation device, equipment and medium |
| CN113763089A (en) * | 2020-10-27 | 2021-12-07 | 北京沃东天骏信息技术有限公司 | Article recommendation method and device and computer-readable storage medium |
| CN112257168A (en) * | 2020-10-30 | 2021-01-22 | 贝壳技术有限公司 | Recommendation method and device for single-space placement item set and computer-readable storage medium |
| CN112257168B (en) * | 2020-10-30 | 2021-08-31 | 贝壳找房(北京)科技有限公司 | Recommendation method and device for single-space placement item set and computer-readable storage medium |
| CN113744016A (en) * | 2020-11-04 | 2021-12-03 | 北京沃东天骏信息技术有限公司 | Object recommendation method and device, equipment and storage medium |
| WO2022095701A1 (en) * | 2020-11-04 | 2022-05-12 | 北京沃东天骏信息技术有限公司 | Method and device for recommending objects, equipment, and storage medium |
| CN113744016B (en) * | 2020-11-04 | 2024-05-24 | 北京沃东天骏信息技术有限公司 | Object recommendation method and device, equipment and storage medium |
| CN113763108A (en) * | 2021-01-29 | 2021-12-07 | 北京京东拓先科技有限公司 | Article pushing method and device, electronic equipment and storage medium |
| CN114022248A (en) * | 2021-11-08 | 2022-02-08 | 中国建设银行股份有限公司 | Product recommendation calculation method, device and equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2022141861A1 (en) | Emotion classification method and apparatus, electronic device, and storage medium | |
| CN111612581A (en) | Method, device and equipment for recommending articles and storage medium | |
| US20220114199A1 (en) | System and method for information recommendation | |
| CN107908740B (en) | Information output method and device | |
| WO2020048051A1 (en) | Financial product recommendation method, server and computer readable storage medium | |
| CN109213936B (en) | Commodity searching method and device | |
| CN111708934A (en) | Knowledge content evaluation method and device, electronic equipment and storage medium | |
| CN110084658B (en) | Method and device for matching articles | |
| CN111400600A (en) | Message pushing method, device, equipment and storage medium | |
| CN111966886A (en) | Object recommendation method, object recommendation device, electronic equipment and storage medium | |
| US11734589B2 (en) | Virtual assistant negotiation system and method thereof | |
| CN111680165A (en) | Information matching method and device, readable storage medium and electronic equipment | |
| CN111861605A (en) | Business object recommendation method | |
| CN111754278A (en) | Article recommendation method and device, computer storage medium and electronic equipment | |
| CN113422986A (en) | Method, apparatus, device, medium, and program product for live room recommendation | |
| CN113742564A (en) | Target resource pushing method and device | |
| CN115827994A (en) | Data processing method, device, equipment and storage medium | |
| CN113362141A (en) | Associated commodity recommendation method, device, medium and electronic equipment | |
| CN115017200A (en) | Search result sorting method and device, electronic equipment and storage medium | |
| CN113722593A (en) | Event data processing method and device, electronic equipment and medium | |
| CN114138976A (en) | Data processing and model training method and device, electronic equipment and storage medium | |
| CN113763075A (en) | Method, device, equipment and medium for pushing articles | |
| CN113313619A (en) | Information pushing method and device, electronic equipment and storage medium | |
| CN114637921B (en) | Item recommendation method, device and equipment based on modeling accidental uncertainty | |
| CN113988923B (en) | Method, device, equipment and storage medium for determining information |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200901 |
|
| RJ01 | Rejection of invention patent application after publication |