CN113672813A

CN113672813A - Article recommendation method, device, terminal and storage medium

Info

Publication number: CN113672813A
Application number: CN202110988088.2A
Authority: CN
Inventors: 陈程; 王贺; 石奕
Original assignee: Wuhan Zhuoer Digital Media Technology Co ltd
Current assignee: Wuhan Zhuoer Digital Media Technology Co ltd
Priority date: 2021-08-26
Filing date: 2021-08-26
Publication date: 2021-11-19

Abstract

The application provides an article recommendation method, an article recommendation device, a terminal and a storage medium, wherein the method comprises the following steps: collecting grading data of a user on an article; dividing the items into a first type of items and a second type of items; based on the scoring data of any user on each item in the first type of items and a predefined time forgetting weight value, obtaining the preference probability of any user on each item in the second type of items; recommending the selected item in the second type of item to the arbitrary user according to the preference probability; and/or obtaining the nearest neighbor user of any user; according to the scoring data of the nearest neighbor user to each article in the second articles, obtaining the prediction score of any user to the second articles; and selecting an item from the second type of items and recommending the selected item to any user according to the prediction score of the any user on the second type of items. The method and the device can assist in recommending the articles and improve the accuracy of recommending the articles.

Description

Article recommendation method, device, terminal and storage medium

Technical Field

The present application relates to the field of computer technologies, and in particular, to a method, an apparatus, a terminal, and a storage medium for recommending an item.

Background

At present, recommendation algorithms are often classified into memory-based recommendation algorithms and model-based recommendation algorithms according to different memory data called in at runtime. The recommendation algorithm based on the memory is to directly call the whole user data during the operation of the algorithm and calculate and recommend on the whole data set in real time; the model-based recommendation algorithm can be divided into a learning stage and a recommendation stage, wherein the learning stage establishes a recommendation model on a data set in an off-line manner, and then only calls the recommendation model into a memory to generate recommendation in the recommendation stage.

Disclosure of Invention

In view of the above, it is desirable to provide an article recommendation method, apparatus, terminal and storage medium, which can assist recommendation and improve the accuracy of article recommendation.

A first aspect of the present application provides an item recommendation method, the method including:

collecting scoring data for each of a plurality of items by a plurality of users;

dividing the plurality of items into a first type of items and a second type of items according to the scoring data of any one user in the plurality of users on each item;

selecting an item from the second category of items using a first method and/or a second method, and recommending the selected item to the any one user, wherein:

the first method includes:

obtaining a first preference probability of any user for each item in the second type of items according to the scoring data of any user for each item in the first type of items; defining a time forgetting weight, and optimizing the first preference probability according to the time forgetting weight to obtain a second preference probability; and

selecting an item from the second category of items according to the second likeness probability, and recommending the selected item to the arbitrary one user;

the second method comprises the following steps:

classifying the plurality of users to obtain the nearest neighbor user of any one user;

according to the scoring data of the nearest neighbor user of any user to each article in the second articles, obtaining the prediction score of any user to each article in the second articles; and

and selecting an item from the second type of items and recommending the selected item to any user according to the prediction score of any user on each item in the second type of items.

Optionally, the first type of items comprises items that have been scored by the any one user, and the second type of items comprises items that have not been scored by the any one user;

wherein the scoring data of each item in the first category of items by any one user comprises: a first rating of each item in the first category of items by the any one user, and a first time at which each item in the first category of items is rated by the any one user.

Optionally, the obtaining a first preference probability of the arbitrary user for each item in the second category of items according to the scoring data of the arbitrary user for each item in the first category of items includes:

calculating a second score for any item in the second category of items, the second score for any item representing an average score for that item by the plurality of users;

defining a score F, and counting the total number l1 of the first type of articles, the number l2 of the first type of articles with the first score less than or equal to F, and the number l3 of the first type of articles with the first score equal to F;

calculating the probability P that the first-class items with the first scores less than or equal to F are liked by any user i according to the total number l1 and the number l2_i(Rating ≦ F), and the probability P that the first type of item with the first score equal to F is liked by said any one user i according to said total number l1 and said number l3_i(Rating＝F)；

According to the probability P_i(Rating ≦ F) and the probability P_i(Rating ═ F) calculationA first likeness probability that a second type of item j having a second score of F is liked by the arbitrary one user i.

Optionally, the defining the time forgetting weight, and optimizing the first preference probability according to the time forgetting weight to obtain a second preference probability includes:

defining a time forgetting function according to the first time when any user i scores any first-class item j, the latest scoring time of any user i and a predefined time gap parameter;

defining the time forgetting weight value based on the time forgetting function and a predefined standard coefficient;

and optimizing the first preference probability according to the time forgetting weight value to obtain the second preference probability.

Optionally, the selecting an item from the second category of items according to the second likeness probability and recommending the selected item to the arbitrary one user includes:

and sorting the values of the second preference probabilities in descending order, selecting the article with the second score F corresponding to the value of the second preference probability in the first preset number, and recommending the selected article to any one user.

Optionally, the classifying the multiple users and the obtaining the nearest neighbor user of the arbitrary user includes:

calculating the similarity between any user A and a user B, wherein the user B represents any other user except the any user A in the plurality of users;

defining a weight factor g;

obtaining an improved similarity between the arbitrary user A and the user B based on the similarity and the weight factor g;

and selecting users corresponding to a preset second number of improved similarity values as the nearest neighbor users of any one user from the improved similarity values in descending order.

Optionally, selecting an item from the second category of items and recommending the selected item to the any one user according to the predictive score of the any one user for each item in the second category of items comprises:

comparing the prediction score to a preset threshold;

and selecting the item corresponding to the prediction score which is larger than the threshold value in the second type of items to recommend to any user.

A second aspect of the present application provides an item recommendation device, the device comprising:

the collection module is used for collecting scoring data of each item in the plurality of items respectively by the plurality of users;

the classification module is used for dividing the plurality of articles into a first type of article and a second type of article according to the scoring data of any one user in the plurality of users on each article;

and the recommending module is used for selecting the articles from the second type of articles by adopting a first method and/or a second method and recommending the selected articles to any user.

A third aspect of the application provides a terminal comprising a processor for implementing the item recommendation method when executing a computer program stored in a memory.

A fourth aspect of the present application provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the item recommendation method.

Compared with the prior art, the item recommendation method, the item recommendation device, the item recommendation terminal and the storage medium deeply research the influence of the user model on the item recommendation accuracy rate, so that the item recommendation accuracy rate is improved, and the item sales volume is increased.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, it is obvious that the drawings in the following description are only embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.

Fig. 1 is a flowchart of an item recommendation method according to an embodiment of the present application.

Fig. 2 is a flowchart of a first method according to an embodiment of the present application.

Fig. 3 is a flowchart of a second method provided in an embodiment of the present application.

Fig. 4 is a structural diagram of an article recommendation device according to a second embodiment of the present application.

Fig. 5 is a schematic structural diagram of a terminal according to a third embodiment of the present application.

Description of the main elements

Article recommendation device	20
		Collection module	201
Classification module	202
		Recommendation module	203
Terminal device	3
		Memory device	31
Processor with a memory having a plurality of memory cells	32
		Communication bus	33
Transceiver	34

The following detailed description will further illustrate the present application in conjunction with the above-described figures.

Detailed Description

In order that the above objects, features and advantages of the present application can be more clearly understood, a detailed description of the present application will be given below with reference to the accompanying drawings and specific embodiments. It should be noted that the embodiments and features of the embodiments of the present application may be combined with each other without conflict.

In the following description, numerous specific details are set forth to provide a thorough understanding of the present application, and the described embodiments are merely a subset of the embodiments of the present application and are not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein in the description of the present application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application.

Example one

Fig. 1 is a flowchart illustrating an item recommendation method according to an embodiment of the present application.

In this embodiment, the item recommendation method may be applied to a terminal, and for a terminal that needs to perform item recommendation, the function provided by the method for applying for item recommendation may be directly integrated on the terminal, or may be run in the terminal in the form of a Software Development Kit (SDK).

As shown in fig. 1, the item recommendation method specifically includes the following steps, and according to different requirements, the order of the steps in the flowchart may be changed, and some steps may be omitted.

Step S1, collecting scoring data for each of a plurality of items by a plurality of users.

Any one of the plurality of users may rate the plurality of items (e.g., movies).

The scoring data includes: the rating of each item by the any one user (e.g., 1 point, 2 points, 3 points), the time at which each item was rated by the any one user (e.g., day of month of year), and the average rating of any one item as rated by the plurality of users.

Wherein the average rating of the any one item evaluated by the plurality of users is the rating of the any one item and/or the number of users who evaluate the any one item.

Step S2, according to the rating data of each item by any one of the users, dividing the items into a first type of item and a second type of item.

The first type of items comprises items scored by the any one user, and the second type of items comprises items not scored by the any one user;

wherein the scoring data of each item in the first category of items by any one user comprises: the scoring of each of the first type of items by the any one user (hereinafter referred to as "first scoring"), and the time at which the any one user scores each of the first type of items (hereinafter referred to as "first time").

And step S3, selecting an item from the second type of items by adopting a first method and/or a second method, and recommending the selected item to any user.

The first method includes steps 4 to 5 as shown in fig. 2.

Step S4, obtaining a first preference probability of the arbitrary user for each item in the second category of items (i.e. the probability that the arbitrary user prefers each item in the second category of items) according to the scoring data of the arbitrary user for each item in the first category of items.

The obtaining a first preference probability of any user for each item in the second category of items according to the scoring data of any user for each item in the first category of items includes:

calculating a second score for any item in the second category of items, the second score for any item representing an average score for the any item by the plurality of users;

calculating the probability P that the first class of articles with the first score less than or equal to F are liked by any user i_i(Rating ≦ F) and a probability P that the first type of item with the first score equal to F is liked by said any one user i_i(Rating ═ F), wherein,

and

calculating a first likeness probability that the second type item j with the second score F is liked by the arbitrary user i

Wherein the content of the first and second substances,

step S5, defining a time forgetting weight value, and optimizing the first preference probability according to the time forgetting weight value to obtain a second preference probability; and selecting an item from the second category of items according to the second preference probability, and recommending the selected item to the arbitrary user.

In this embodiment, a time forgetting function is defined:

wherein, t_i,jRepresents a first time at which any one user i scores any one first type item j, td represents a time at which the any one user i scores the last time, and T represents a predefined time gap parameter (e.g., any constant of 0 to 1).

Based on the time forgetting function, defining the time forgetting weight value as:

where w represents a predefined normalization coefficient (e.g., any constant of 0 to 1).

Optimizing a second preference probability F obtained by the first preference probability according to the time forgetting weight_i,jThe formula of (1) is:

the selecting an item from the second category of items according to the second likeness probability and recommending the selected item to the arbitrary one user includes:

the second preference probability F_i,jIs sorted in order from large to small, a first predetermined number (e.g., 8) of second preference probabilities F arranged in front is selected_i,jThe selected item is recommended to the arbitrary one of the users, with the items having the second scores F corresponding to the values of (a).

It should be noted that the recommendation may be to push the picture information of the item to the user homepage of any one of the users.

The second method includes steps 6 to 8 as shown in fig. 3.

The second method recommends a commodity for the arbitrary one user using a user-based nearest neighbor recommendation (user-based neighbor recommendation) concept.

The nearest neighbor recommendation idea based on the user comprises the following steps:

firstly, using collected scoring data and ID of any user as input, finding out other users with similar preference to any user in the past, and calling the found users as peer-to-peer users or nearest neighbor users of any user;

then, for any item p in the second type of items, obtaining a prediction score of any user for any item p in the second type of items by using the nearest neighbor user of any user;

and finally, selecting the items with the prediction scores higher than a preset threshold value from the second type of items and recommending the items to any user.

The premises/assumptions of the user-based nearest neighbor recommendation method include:

if any user scores a certain item higher in the past, the any user will score the certain item higher in the future, and the score of the any user on the certain item will not change greatly over time.

The calculation mode of the nearest neighbor recommendation method based on the user comprises the following steps:

the similarity (or called Correlation) between two users is calculated by using a Pearson Correlation Coefficient (Pearson Correlation Coefficient), the range of the Pearson Correlation Coefficient is [ -1, +1], wherein the Pearson Correlation Coefficient is-1, which indicates that the two users have strong negative Correlation, the Pearson Correlation Coefficient is +1, which indicates that the two users have strong positive Correlation, and the Pearson Correlation Coefficient is 0, which indicates that the two users have no Correlation.

Step S6, classifying the users, and obtaining the nearest neighbor user of any user.

Calculating the similarity sim (A, B) between any user A and a user B, wherein the user B represents any other user except the user A, and the formula used is as follows:

wherein R is_A,jRepresenting a first rating of any of the items of the first type j by said any one user a,

represents the average of the scores, R, of any one of the users A for all items of the first type_B,jRepresents the user B's score for said any one item of the first type j,

representing the average of the scores of user B for all items of the first type and items of the second type, I_ABA set of items of a first type that represent said any one of user a and user B having a common rating.

Defining weight factors

Where θ represents a preset adjustable parameter.

Obtaining the improved similarity sim' (A, B) between the arbitrary user A and the user B by using the weight factor g, wherein the formula is as follows: sim' (a, B) ═ g × sim (a, B).

Selecting users corresponding to a preset second number (for example, 4) of similarity sim '(A, B) values as the nearest neighbor users of the arbitrary user from the similarity sim' (A, B) values in descending order, and recording the set of the nearest neighbor users of the arbitrary user A as N_A。

Step S7, obtaining the prediction score of each item in the second category of items by the any user according to the score data of the nearest neighbor user of the any user on each item in the second category of items.

According to the scoring data of the nearest neighbor user of any user to each item in the second type of items, the formula for obtaining the prediction score of any user to each item in the second type of items is as follows:

wherein Q (A, k) represents the predictive score of any user A on any second type item k, and N_ARepresenting the set of nearest neighbors, R_B,kRepresenting the score of the nearest user B on the second type of item k.

Step S8, selecting an item from the second type of item and recommending the selected item to the any user according to the prediction score of the any user for each item in the second type of item.

Comparing the prediction score Q (a, k) to a preset threshold (e.g., 2.5);

and selecting the item corresponding to the prediction score Q (A, k) which is greater than the threshold value in the second type of items to recommend to the any user.

In conclusion, the item recommendation method provided by the application deeply studies the influence of the user model on the item recommendation accuracy rate, so that the item recommendation accuracy rate is improved, and the sales volume of the items is increased.

Example two

Fig. 3 is a structural diagram of an article recommendation device according to a second embodiment of the present invention.

In some embodiments, the item recommendation device 20 may include a plurality of functional modules composed of program code segments. The program code of each program segment in the item recommendation device 20 may be stored in a memory of the terminal and executed by the at least one processor to perform (see detailed description of fig. 1) the functions of item recommendation.

In this embodiment, the item recommendation device 20 may be divided into a plurality of functional modules according to the functions performed by the item recommendation device. The functional module may include: a collection module 201, a classification module 202 and a recommendation module 203. The module referred to herein is a series of computer program segments capable of being executed by at least one processor and capable of performing a fixed function and is stored in memory. In the present embodiment, the functions of the modules will be described in detail in the following embodiments.

A collecting module 201, configured to collect scoring data of each of the plurality of items by the plurality of users respectively.

Collecting rating data for the plurality of items (e.g., movies) by any one of the plurality of users.

A classification module 202, configured to classify the multiple items into a first type of item and a second type of item according to the scoring data of each item by any one of the multiple users.

And the recommending module 203 is used for selecting the items from the second type of items by adopting a first method and/or a second method and recommending the selected items to any user.

The first method includes:

selecting an item from the second category of items according to the second likeness probability, and recommending the selected item to the arbitrary one of the users.

and

Wherein the content of the first and second substances,

defining a time forgetting weight, wherein optimizing the first preference probability according to the time forgetting weight to obtain a second preference probability comprises:

defining a time forgetting function:

wherein, t_i,jRepresents a first time at which any one user i scores any one first-type item j, td represents a time at which the any one user i scores the last time, and T represents a predefined time gap parameter (e.g., any constant of 0 to 1);

where w represents a predefined normalization coefficient (e.g., any constant of 0 to 1);

The second method comprises the following steps:

The classifying the plurality of users and the obtaining the nearest neighbor user of any one user comprises:

Defining weight factors

Where θ represents a preset adjustable parameter.

The formula used for obtaining the prediction score of any user for each item in the second type of items according to the score data of the nearest neighbor user of any user for each item in the second type of items is as follows:

Selecting an item from the second category of items and recommending the selected item to the any user based on the predictive scores of the any user for each item in the second category of items comprises:

comparing the prediction score Q (a, k) to a preset threshold (e.g., 2.5);

EXAMPLE III

Fig. 5 is a schematic structural diagram of a terminal according to a third embodiment of the present invention. In the preferred embodiment of the present invention, the terminal 3 includes a memory 31, at least one processor 32, at least one communication bus 33, and a transceiver 34.

It will be appreciated by those skilled in the art that the configuration of the terminal shown in fig. 5 is not limiting to the embodiments of the present invention, and may be a bus-type configuration or a star-type configuration, and the terminal 3 may include more or less hardware or software than those shown, or a different arrangement of components.

In some embodiments, the terminal 3 includes a terminal capable of automatically performing numerical calculation and/or information processing according to preset or stored instructions, and the hardware includes but is not limited to a microprocessor, an application specific integrated circuit, a programmable gate array, a digital processor, an embedded device, and the like. The terminal 3 may further include a client device, which includes, but is not limited to, any electronic product capable of performing human-computer interaction with a client through a keyboard, a mouse, a remote controller, a touch panel, or a voice control device, for example, a personal computer, a tablet computer, a smart phone, a digital camera, and the like.

It should be noted that the terminal 3 is only an example, and other existing or future electronic products, such as those that can be adapted to the present invention, should also be included in the scope of the present invention, and are included herein by reference.

In some embodiments, the memory 31 is used for storing program codes and various data, such as devices installed in the terminal 3, and realizes high-speed and automatic access to programs or data during the operation of the terminal 3. The Memory 31 includes a Read-Only Memory (ROM), a Random Access Memory (RAM), a Programmable Read-Only Memory (PROM), an Erasable Programmable Read-Only Memory (EPROM), a One-time Programmable Read-Only Memory (OTPROM), an electronically Erasable rewritable Read-Only Memory (EEPROM), a Compact Disc Read-Only Memory (CD-ROM) or other optical Disc Memory, a magnetic disk Memory, a tape Memory, or any other medium readable by a computer that can be used to carry or store data.

In some embodiments, the at least one processor 32 may be composed of an integrated circuit, for example, a single packaged integrated circuit, or may be composed of a plurality of integrated circuits packaged with the same or different functions, including one or more Central Processing Units (CPUs), microprocessors, digital Processing chips, graphics processors, and combinations of various control chips. The at least one processor 32 is a Control Unit (Control Unit) of the terminal 3, connects various components of the entire terminal 3 using various interfaces and lines, and executes various functions of the terminal 3 and processes data by running or executing programs or modules stored in the memory 31 and calling data stored in the memory 31.

In some embodiments, the at least one communication bus 33 is arranged to enable connection communication between the memory 31 and the at least one processor 32 or the like.

Although not shown, the terminal 3 may further include a power supply (such as a battery) for supplying power to various components, and preferably, the power supply may be logically connected to the at least one processor 32 through a power management device, so as to implement functions of managing charging, discharging, and power consumption through the power management device. The power supply may also include any component of one or more dc or ac power sources, recharging devices, power failure detection circuitry, power converters or inverters, power status indicators, and the like. The terminal 3 may further include various sensors, a bluetooth module, a Wi-Fi module, and the like, which are not described herein again.

It is to be understood that the described embodiments are for purposes of illustration only and that the scope of the appended claims is not limited to such structures.

The integrated unit implemented in the form of a software functional module may be stored in a computer-readable storage medium. The software functional module is stored in a storage medium and includes several instructions to enable a computer device (which may be a personal computer, a terminal, or a network device) or a processor (processor) to execute parts of the methods according to the embodiments of the present invention.

In a further embodiment, in conjunction with fig. 4, the at least one processor 32 may execute operating means of the terminal 3 as well as installed various types of applications, program codes, etc., such as the various modules described above.

The memory 31 has program code stored therein, and the at least one processor 32 can call the program code stored in the memory 31 to perform related functions. For example, the respective modules illustrated in fig. 4 are program codes stored in the memory 31 and executed by the at least one processor 32, thereby implementing the functions of the respective modules.

In one embodiment of the invention, the memory 31 stores a plurality of instructions that are executed by the at least one processor 32 to implement all or a portion of the steps of the method of the invention.

Specifically, the method for implementing the instruction by the at least one processor 32 may refer to the description of the relevant steps in the embodiments corresponding to fig. 1 to fig. 3, which is not repeated herein.

In the embodiments provided in the present invention, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is only one logical functional division, and other divisions may be realized in practice.

The modules described as separate parts may or may not be physically separate, and parts displayed as modules may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment.

In addition, functional modules in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional module.

It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned. Furthermore, it is obvious that the word "comprising" does not exclude other elements or that the singular does not exclude the plural. A plurality of units or means recited in the apparatus claims may also be implemented by one unit or means in software or hardware. The terms first, second, etc. are used to denote names, but not any particular order.

Finally, it should be noted that the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting, and although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An item recommendation method, characterized in that the method comprises:

the first method includes:

the second method comprises the following steps:

2. The item recommendation method according to claim 1, wherein the first type of item comprises an item that has been scored by the any one user, and the second type of item comprises an item that has not been scored by the any one user;

3. The item recommendation method according to claim 2, wherein the obtaining a first preference probability for each item in the second category of items by the any user according to the scoring data for each item in the first category of items by the any user comprises:

According to the probability P_i(Rating ≦ F) and the probability P_i(Rating ═ F) a first likeliness probability that a second item j with a second score of F is liked by said any one user i is calculated.

4. The item recommendation method according to claim 3, wherein the defining a time forgetting weight, and the optimizing the first preference probability according to the time forgetting weight to obtain a second preference probability comprises:

5. The item recommendation method according to claim 4, wherein said selecting an item from said second category of items according to said second likeness probability and recommending the selected item to said any one user comprises:

6. The item recommendation method according to claim 2, wherein said classifying the plurality of users and obtaining a nearest neighbor user of the arbitrary user comprises:

defining a weight factor g;

7. The item recommendation method according to claim 1, wherein selecting and recommending the selected item to the any user from the second category of items based on the predictive score of the any user for each item in the second category of items comprises:

comparing the prediction score to a preset threshold;

8. An item recommendation device, the device comprising:

9. A terminal, characterized in that the terminal comprises a processor for implementing the item recommendation method according to any one of claims 1 to 7 when executing a computer program stored in a memory.

10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out an item recommendation method according to any one of claims 1 to 7.