WO2019085120A1 - Collaborative filtering recommendation method, electronic device, and computer readable storage medium - Google Patents

Abstract

A collaborative filtering recommendation method. The method comprises the steps of: calculating similarities of a target user to different designated users according to a user tag correlation matrix, and selecting a first preset number of designated users as a nearest neighbor set of the target user according to a descending order of the similarities (S31); selecting, from user tags of the nearest neighbor set, tags that are not used by the target user, and using same as candidate tags of the target user (S32); calculating a correlation of the target user to each candidate tag, and selecting a second preset number of candidate tags as newly-added tags of the target user according to a descending order of the correlations (S33); and recommending items related to original tags and the newly-added tags to the target user according to the original tags and the newly-added tags of the target user (S34). The recommendation efficiency can be improved.

Description

Collaborative filtering recommendation method, electronic device and computer readable storage medium

This application claims the priority of the Chinese Patent Application filed on November 1, 2017, the Chinese Patent Office, Application No. 201711059396.7, entitled "Collaborative Filtering Recommendation Method, Electronic Device, and Computer Readable Storage Media", the entire contents of which are hereby incorporated by reference. The citation is incorporated in the application.

Technical field

The present application relates to the field of computer information technology, and in particular, to a collaborative filtering recommendation method, an electronic device, and a computer readable storage medium.

Background technique

At present, the traditional collaborative filtering recommendation methods mainly include user-based collaborative filtering recommendation and project-based collaborative filtering recommendation. However, in the case where the scoring matrix is sparse, the traditional collaborative filtering recommendation algorithm may not be able to perform similarity calculation. Moreover, as the user grows, the amount of calculation increases linearly, and the scalability is not ideal. Therefore, the design of the collaborative filtering recommendation method in the prior art is not reasonable enough and needs to be improved.

Summary of the invention

In view of this, the present application proposes a collaborative filtering recommendation method, an electronic device, and a computer readable storage medium, and solves the sparseness and interest of the traditional collaborative filtering method scoring matrix by introducing a user-tag correlation matrix and a tag-item correlation matrix. The single problem of the model reduces the scale of the scoring matrix and improves the efficiency of the algorithm.

First, in order to achieve the above object, the present application provides an electronic device, where the electronic device includes a memory and a processor, where the memory stores a collaborative filtering recommendation system that can be run on the processor, and the collaborative filtering recommendation The system implements the following steps when executed by the processor:

Calculating the similarity between the target user and the different designated users according to the user-tag correlation matrix, and selecting the first predetermined number of designated users as the nearest neighbor set of the target user according to the order of similarity from high to low;

Selecting, from the user tags of the nearest neighbor set, a tag that is not used by the target user as a candidate tag of the target user;

Calculating a correlation between the target user and each candidate tag, and selecting a second predetermined number of candidate tags as the new tag of the target user according to the order of relevance from highest to lowest; and

According to the original label and the newly added label of the target user, items related to the original label and the newly added label are recommended to the target user.

Preferably, the user-tag correlation matrix adopts a two-dimensional matrix, each row of the two-dimensional matrix represents a user, and each column represents a label, and the two-dimensional matrix stores a correlation between all users and all tags;

The similarity between the target user and different designated users is calculated by using a first calculation formula. The first calculation formula is set to Equation 1:

In Equation 1, Sa _{a, s} represents the similarity between the target user u _a and the specified user u _s , and T _{a, s} represents the label used by the target user u _a and the specified user u _s , and r _{a, t} represents the target The correlation between the user u _a and the tag t, r _s,t represents the correlation between the specified user u _s and the tag t,

The average of the correlation between the user u _a and all tags,

Represents the average of the correlation of user u _s with all tags.

Preferably, the correlation between the target user and each candidate tag is calculated by using a second calculation formula, and the second calculation formula is set to formula 2:

In Equation 2, P _a,k represents the correlation of the target user u _a with the candidate tag t _k , U _K represents the user set using the tag t _k , U _N represents all users of the nearest neighbor set, Sa _{, u} Representing the similarity between the target user u _a and the specified user u _u , r _u, represents the correlation between the specified user u _u and the candidate tag t _k .

Preferably, the recommending the item related to the original label and the newly added label to the target user comprises:

Selecting a label from the original label and the newly added label of the target user to obtain a set of items marked by the selected label, and calculating a correlation between the selected label and each item in the item set, in descending order of relevance. , selecting a third predetermined number of items to recommend to the target user.

Preferably, the correlation between the selection label and each item in the project set is calculated by using a third calculation formula, and the third calculation formula is set to formula 3:

In Equation 3, t ₁ represents an original tag of the target user and a tag in the newly added tag set T _N , i represents an item in the item set I _t1 marked by the tag t ₁ , and relat(t ₁ , i) represents The correlation of the tag t ₁ with the item i, countUser(t ₁ , i) represents the number of users associated with item i and tag t ₁ , and countUser(t ₁ , j) represents the number of users associated with item j and tag t ₁ .

In addition, to achieve the above object, the present application further provides a collaborative filtering recommendation method, which is applied to an electronic device, and the method includes:

Calculating the similarity between the target user and the different designated users according to the user-tag correlation matrix, and selecting the first predetermined number of designated users as the nearest neighbor set of the target user according to the order of similarity from high to low;

Selecting, from the user tags of the nearest neighbor set, a tag that is not used by the target user as a candidate tag of the target user;

Calculate the relevance of the target user to each candidate tag, in descending order of relevance, Selecting a second predetermined number of candidate tags as the new tag of the target user; and

According to the original label and the newly added label of the target user, items related to the original label and the newly added label are recommended to the target user.

Preferably, the user-tag correlation matrix adopts a two-dimensional matrix, each row of the two-dimensional matrix represents a user, and each column represents a label, and the two-dimensional matrix stores a correlation between all users and all tags;

The similarity between the target user and different specified users is calculated by using a first calculation formula, and the first calculation formula is set to formula 1:

In Equation 1, Sa _{a, s} represents the similarity between the target user u _a and the specified user u _s , and T _{a, s} represents the label used by the target user u _a and the specified user u _s , and r _{a, t} represents the target The correlation between the user u _a and the tag t, r _s,t represents the correlation between the specified user u _s and the tag t,

The average of the correlation between the user u _a and all tags,

Represents the average of the correlation of user u _s with all tags.

Preferably, the correlation between the target user and each candidate tag is calculated by using a second calculation formula, and the second calculation formula is set to formula 2:

In Equation 2, P _a,k represents the correlation of the target user u _a with the candidate tag t _k , U _K represents the user set using the tag t _k , U _N represents all users of the nearest neighbor set, Sa _{, u} Representing the similarity between the target user u _a and the specified user u _u , r _u, represents the correlation between the specified user u _u and the candidate tag t _k .

Preferably, the recommending the item related to the original label and the newly added label to the target user comprises:

Selecting a label from the original label and the newly added label of the target user to obtain a set of items marked by the selected label, and calculating a correlation between the selected label and each item in the item set, in descending order of relevance. , selecting a third predetermined number of items to recommend to the target user;

The correlation between the selection label and each item in the project set is calculated by using a third calculation formula, and the third calculation formula is set to formula 3:

In Equation 3, t ₁ represents an original tag of the target user and a tag in the newly added tag set T _N , i represents an item in the item set I _t1 marked by the tag t ₁ , and relat(t ₁ , i) represents The correlation of the tag t ₁ with the item i, countUser(t ₁ , i) represents the number of users associated with item i and tag t ₁ , and countUser(t ₁ , j) represents the number of users associated with item j and tag t ₁ .

Further, to achieve the above object, the present application further provides a computer readable storage medium, The computer readable storage medium stores a collaborative filtering recommendation system executable by at least one processor to cause the at least one processor to perform the steps of the collaborative filtering recommendation method as described above.

Compared with the prior art, the electronic device, the collaborative filtering recommendation method and the computer readable storage medium proposed by the present application solve the traditional collaborative filtering method scoring matrix by introducing a user-tag correlation matrix and a tag-item correlation matrix. The sparseness and interest model are single, which reduces the scale of the scoring matrix, improves the efficiency of the algorithm, and enhances the scalability of the algorithm. The user label-based collaborative filtering recommendation effect used in this application is superior to the traditional collaborative filtering method.

DRAWINGS

1 is a schematic diagram of an optional hardware architecture of an electronic device of the present application;

2 is a schematic diagram of a program module of an embodiment of a collaborative filtering recommendation system in an electronic device of the present application;

FIG. 3 is a schematic diagram of an implementation process of an embodiment of a collaborative filtering recommendation method according to the present application.

Reference mark:

Electronic equipment	2
Memory	twenty one
processor	twenty two
Network Interface	twenty three
Collaborative filtering recommendation system	20
Calculation module	201
Selection module	202
Recommended module	203
Process step	S31-S34

The implementation, functional features and advantages of the present application will be further described with reference to the accompanying drawings.

Detailed ways

In order to make the objects, technical solutions, and advantages of the present application more comprehensible, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It is understood that the specific embodiments described herein are merely illustrative of the application and are not intended to be limiting. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of the present application without departing from the inventive scope are the scope of the present application.

It should be noted that the descriptions of "first", "second" and the like in the present application are for the purpose of description only, and are not to be construed as indicating or implying their relative importance or implicitly indicating the number of technical features indicated. . Thus, features defining "first" or "second" may include at least one of the features, either explicitly or implicitly. In addition, the technical solutions between the various embodiments may be combined with each other, but must be based on the realization of those skilled in the art, and when the combination of the technical solutions is contradictory or impossible to implement, it should be considered that the combination of the technical solutions does not exist. Nor is it within the scope of protection required by this application.

It is further to be understood that the term "comprises", "comprises" or any other variations thereof is intended to encompass a non-exclusive inclusion, such that a process, method, article, or device that comprises a And includes other elements not explicitly listed, or elements that are inherent to such a process, method, article, or device. An element that is defined by the phrase "comprising a ..." does not exclude the presence of additional equivalent elements in the process, method, item, or device that comprises the element.

First of all, the present application proposes an electronic device 2.

Referring to FIG. 1, it is a schematic diagram of an optional hardware architecture of the electronic device 2 of the present application. In this embodiment, the electronic device 2 may include, but is not limited to, a memory 21, a processor 22, and a network interface 23 that can communicate with each other through a system bus. It is pointed out that FIG. 1 only shows the electronic device 2 with the components 21-23, but it should be understood that not all illustrated components are required to be implemented, and more or fewer components may be implemented instead.

The electronic device 2 may be a computing device such as a rack server, a blade server, a tower server, or a rack server. The electronic device 2 may be an independent server or a server cluster composed of multiple servers. .

The memory 21 includes at least one type of readable storage medium including a flash memory, a hard disk, a multimedia card, a card type memory (eg, SD or DX memory, etc.), a random access memory (RAM), a static Random access memory (SRAM), read only memory (ROM), electrically erasable programmable read only memory (EEPROM), programmable read only memory (PROM), magnetic memory, magnetic disk, optical disk, and the like. In some embodiments, the memory 21 may be an internal storage unit of the electronic device 2, such as a hard disk or memory of the electronic device 2. In other embodiments, the memory 21 may also be an external storage device of the electronic device 2, such as a plug-in hard disk equipped on the electronic device 2, a smart memory card (SMC), and a secure digital device. (Secure Digital, SD) card, flash card, etc. Of course, the memory 21 may also include both an internal storage unit of the electronic device 2 and an external storage device thereof. In this embodiment, the memory 21 is generally used to store an operating system installed in the electronic device 2 and various types of application software, such as program code of the collaborative filtering recommendation system 20. Further, the memory 21 can also be used to temporarily store various types of data that have been output or are to be output.

The processor 22 may be a Central Processing Unit (CPU), controller, microcontroller, microprocessor, or other data processing chip in some embodiments. This treatment The device 22 is typically used to control the overall operation of the electronic device 2, such as performing control and processing associated with data interaction or communication with the electronic device 2. In this embodiment, the processor 22 is configured to run program code or process data stored in the memory 21, such as running the collaborative filtering recommendation system 20 and the like.

The network interface 23 may comprise a wireless network interface or a wired network interface, which is typically used to establish a communication connection between the electronic device 2 and other electronic devices. For example, the network interface 23 is configured to connect the electronic device 2 to an external data platform through a network, and establish a data transmission channel and a communication connection between the electronic device 2 and an external data platform. The network may be an intranet, an Internet, a Global System of Mobile communication (GSM), a Wideband Code Division Multiple Access (WCDMA), a 4G network, or a 5G network. Wireless or wired networks such as network, Bluetooth, Wi-Fi, etc.

So far, the application environment of the various embodiments of the present application and the hardware structure and functions of related devices have been described in detail. Hereinafter, various embodiments of the present application will be proposed based on the above-described application environment and related devices.

Referring to FIG. 2, it is a program module diagram of an embodiment of the collaborative filtering recommendation system 20 in the electronic device 2 of the present application. In this embodiment, the collaborative filtering recommendation system 20 may be divided into one or more program modules, the one or more program modules being stored in the memory 21 and being processed by one or more processors ( This embodiment is executed by the processor 22) to complete the application. For example, in FIG. 2, the collaborative filtering recommendation system 20 can be divided into a computing module 201, a selection module 202, and a recommendation module 203. The program module referred to in the present application refers to a series of computer program instruction segments capable of performing a specific function, and is more suitable than the program to describe the execution process of the collaborative filtering recommendation system 20 in the electronic device 2. The function of each program module 201-203 will be described in detail below.

The calculating module 201 is configured to calculate a similarity between the target user and different specified users according to the user-tag correlation matrix, and select the first predetermined number of designated users according to the order of similarity from high to low (if the similarity is high) The first 10 specified users) are the closest neighbor collection of the target user.

Preferably, in this embodiment, the user-tag correlation matrix adopts a two-dimensional matrix, each row of the two-dimensional matrix represents a user, and each column represents a tag (such as a property insurance user tag, etc.). Further, the two-dimensional matrix stores the correlation between all users and all tags.

Preferably, in this embodiment, the similarity between the target user and different specified users is calculated by using a first calculation formula, wherein the first calculation formula may be set as shown in the following formula 1.

Wherein, the Sa _{, s} represents a similarity between the target user u _a and the specified user u _s , and T _{a, s} represents a label used by the target user u _a and the specified user u _s , and r _{a, t} represents the target The correlation between the user u _a and the tag t, r _s,t represents the correlation between the specified user u _s and the tag t,

The average of the correlation between the user u _a and all tags,

Represents the average of the correlation of user u _s with all tags.

The selecting module 202 is configured to select, from the user tags of the nearest neighbor set, a tag that is not used by the target user, as a candidate tag of the target user (recorded as a set).

The calculation module 201 is further configured to calculate a correlation between the target user and each candidate tag, and select a second predetermined number of candidate tags according to a high-to-low correlation (for example, the first three related correlations) Candidate tag) as a new tag for this target user.

Preferably, in this embodiment, the correlation between the target user and each candidate tag is calculated by using a second calculation formula, wherein the second calculation formula may be set as shown in the following formula 2.

Wherein, the P _a,k represents the correlation between the target user u _a and the candidate tag t _k , U _K represents the user set using the tag t _k , and U _N represents all users of the nearest neighbor set (ie the first reservation) a specified number of users, N = first predetermined number), Sa _{, u} represents the similarity between the target user u _a and the specified user u _u , and r _{u, k} represents the relationship between the specified user u _u and the candidate tag t _k Correlation (ie, assigning the user u _{u to} use the weight of the candidate tag t _k ).

The recommendation module 203 is configured to recommend an item related to the original label and the newly added label to the target user according to the original label and the newly added label of the target user. The original label of the target user may be a label of the target user originally stored in the user-tag correlation matrix.

Preferably, in this embodiment, the recommending the item related to the original label and the newly added label to the target user includes the following steps:

Selecting a label from the original label and the newly added label of the target user, obtaining a set of items marked by the selected label, and calculating a correlation between the selected label and each item in the item set (getting a label-item correlation matrix), According to the order of relevance from high to low, the third predetermined number of items (such as the top 3 items with higher relevance) are selected and recommended to the target user.

Further, in this embodiment, the correlation between the selection label and each item in the item set is calculated by using a third calculation formula, wherein the third calculation formula may be set as shown in the following formula 3.

Where t ₁ represents an original label of the target user and a label in the newly added label set T _N (ie, t ₁ ∈T _N ), and i represents an item in the item set I _t1 marked by the label t ₁ , t ₁ , i) represents the correlation of the tag t ₁ with the item i, countUser(t ₁ , i) represents the number of users associated with the item i and the tag t ₁ , and countUser(t ₁ , j) represents the item j and the tag t ₁ The number of related users, the right denominator part of Equation 3 represents the number of all users associated with the tag t ₁ in the item set I _t1 .

It should be noted that the traditional user-based collaborative filtering and project-based collaborative filtering have problems such as sparse scoring matrix, single interest model, and cold start. This application introduces user tags into collaborative filtering through user-tag correlation matrices and tags. - The project correlation matrix is collaboratively filtered, thus eliminating the traditional user-project matrix model and using the tags to divide the project to make up for the shortcomings of traditional collaborative filtering.

Through the above-mentioned program modules 201-203, the collaborative filtering recommendation system 20 proposed by the present application solves the problem of sparse scoring matrix and single interest model of the traditional collaborative filtering method by introducing a user-tag correlation matrix and a tag-item correlation matrix. The scale of the scoring matrix is reduced, the efficiency of the algorithm is improved, and the scalability of the algorithm is enhanced. The collaborative filtering recommendation effect based on user tags used in the present application is superior to the traditional collaborative filtering method.

In addition, the present application also proposes a collaborative filtering recommendation method.

Referring to FIG. 3, it is a schematic flowchart of an implementation process of an embodiment of the collaborative filtering recommendation method of the present application. In this embodiment, the order of execution of the steps in the flowchart shown in FIG. 3 may be changed according to different requirements, and some steps may be omitted.

Step S31, calculating the similarity between the target user and different designated users according to the user-tag correlation matrix, and selecting the first predetermined number of designated users according to the order of similarity from high to low (eg, the top 10 designations with higher similarity) User) as the closest neighbor set for the target user.

Preferably, in this embodiment, the user-tag correlation matrix adopts a two-dimensional matrix, each row of the two-dimensional matrix represents a user, and each column represents a tag (such as a property insurance user tag, etc.). Further, the two-dimensional matrix stores the correlation between all users and all tags.

Preferably, in this embodiment, the similarity between the target user and different specified users is calculated by using a first calculation formula, wherein the first calculation formula may be set as shown in the following formula 1.

Wherein, the Sa _{, s} represents a similarity between the target user u _a and the specified user u _s , and T _{a, s} represents a label used by the target user u _a and the specified user u _s , and r _{a, t} represents the target The correlation between the user u _a and the tag t, r _s,t represents the correlation between the specified user u _s and the tag t,

The average of the correlation between the user u _a and all tags,

Represents the average of the correlation of user u _s with all tags.

Step S32: Select a label that is not used by the target user from the user labels of the nearest neighbor set as a candidate label of the target user (recorded as a set).

Step S33, calculating a correlation between the target user and each candidate tag, and selecting a second predetermined number of candidate tags (such as the top 3 candidate tags with higher correlation) as the target according to the order of relevance from high to low. User's new label.

Preferably, in this embodiment, the correlation between the target user and each candidate tag is calculated by using a second calculation formula, wherein the second calculation formula may be set as shown in the following formula 2.

Wherein, the P _a,k represents the correlation between the target user u _a and the candidate tag t _k , U _K represents the user set using the tag t _k , and U _N represents all users of the nearest neighbor set (ie the first reservation) a specified number of users, N = first predetermined number), Sa _{, u} represents the similarity between the target user u _a and the specified user u _u , and r _{u, k} represents the relationship between the specified user u _u and the candidate tag t _k Correlation (ie, assigning the user u _{u to} use the weight of the candidate tag t _k ).

Step S34, recommending items related to the original label and the newly added label to the target user according to the original label and the newly added label of the target user. The original label of the target user may be a label of the target user originally stored in the user-tag correlation matrix.

Preferably, in this embodiment, the recommending the item related to the original label and the newly added label to the target user includes the following steps:

Selecting a label from the original label and the newly added label of the target user, obtaining a set of items marked by the selected label, and calculating a correlation between the selected label and each item in the item set (getting a label-item correlation matrix), According to the order of relevance from high to low, the third predetermined number of items (such as the top 3 items with higher relevance) are selected and recommended to the target user.

Further, in this embodiment, the correlation between the selection label and each item in the item set is calculated by using a third calculation formula, wherein the third calculation formula may be set as shown in the following formula 3.

Where t ₁ represents an original label of the target user and a label in the newly added label set T _N (ie, t ₁ ∈T _N ), and i represents an item in the item set I _t1 marked by the label t ₁ , t ₁ , i) represents the correlation of the tag t ₁ with the item i, countUser(t ₁ , i) represents the number of users associated with the item i and the tag t ₁ , and countUser(t ₁ , j) represents the item j and the tag t ₁ The number of related users, the right denominator part of Equation 3 represents the number of all users associated with the tag t ₁ in the item set I _t1 .

It should be noted that the traditional user-based collaborative filtering and project-based collaborative filtering have problems such as sparse scoring matrix, single interest model, and cold start. This application introduces user tags into collaborative filtering through user-tag correlation matrices and tags. - The project correlation matrix is collaboratively filtered, thus eliminating the traditional user-project matrix model and using the tags to divide the project to make up for the shortcomings of traditional collaborative filtering.

Through the above steps S31-S34, the collaborative filtering recommendation method proposed by the present application solves the problem of sparse scoring matrix and single interest model of the traditional collaborative filtering method by introducing a user-tag correlation matrix and a label-item correlation matrix, and reduces the problem. The scale of the scoring matrix improves the efficiency of the algorithm and enhances the scalability of the algorithm. The user label-based collaborative filtering recommendation effect used in this application is superior to the traditional collaborative filtering method.

Further, in order to achieve the above object, the present application further provides a computer readable storage medium (such as a ROM/RAM, a magnetic disk, an optical disk), where the computer readable storage medium stores a collaborative filtering recommendation system 20, and the collaborative filtering The recommendation system 20 can be executed by at least one processor 22 to cause the at least one processor 22 to perform the steps of the collaborative filtering recommendation method as described above.

Through the description of the above embodiments, those skilled in the art can clearly understand that the foregoing embodiment method can be implemented by means of software plus a necessary general hardware platform, and can also be implemented by hardware, but in many cases, the former is A better implementation. Based on such understanding, the technical solution of the present application, which is essential or contributes to the prior art, may be embodied in the form of a software product stored in a storage medium (such as ROM/RAM, disk, The optical disc includes a number of instructions for causing a terminal device (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the methods described in various embodiments of the present application.

The preferred embodiments of the present application have been described above with reference to the drawings, and are not intended to limit the scope of the application. The serial numbers of the embodiments of the present application are merely for the description, and do not represent the advantages and disadvantages of the embodiments. Additionally, although logical sequences are shown in the flowcharts, in some cases the steps shown or described may be performed in a different order than the ones described herein.

A person skilled in the art can implement the present application in various variants without departing from the scope and spirit of the present application. For example, the features as one embodiment can be used in another embodiment to obtain another embodiment. The equivalent structure or equivalent process transformations made by the present specification and the contents of the drawings, or directly or indirectly applied to other related technical fields, are all included in the scope of patent protection of the present application.

Claims (20)

1. An electronic device, comprising: a memory and a processor, wherein the memory stores a collaborative filtering recommendation system operable on the processor, the collaborative filtering recommendation system being The following steps are implemented during execution:

   Calculating the similarity between the target user and the different designated users according to the user-tag correlation matrix, and selecting the first predetermined number of designated users as the nearest neighbor set of the target user according to the order of similarity from high to low;

   Selecting, from the user tags of the nearest neighbor set, a tag that is not used by the target user as a candidate tag of the target user;

   Calculating a correlation between the target user and each candidate tag, and selecting a second predetermined number of candidate tags as the new tag of the target user according to the order of relevance from highest to lowest; and

   According to the original label and the newly added label of the target user, items related to the original label and the newly added label are recommended to the target user.
2. The electronic device according to claim 1, wherein the user-tag correlation matrix adopts a two-dimensional matrix, each row of the two-dimensional matrix represents a user, and each column represents a label, and the two-dimensional matrix stores Correlation between all users and all tags;

   The similarity between the target user and different specified users is calculated by using a first calculation formula, and the first calculation formula is set to formula 1:

   

   In Equation 1, Sa _{a, s} represents the similarity between the target user u _a and the specified user u _s , and T _{a, s} represents the label used by the target user u _a and the specified user u _s , and r _{a, t} represents the target The correlation between the user u _a and the tag t, r _s,t represents the correlation between the specified user u _s and the tag t,

   

   The average of the correlation between the user u _a and all tags,

   

   Represents the average of the correlation of user u _s with all tags.
3. The electronic device according to claim 1, wherein the correlation between the target user and each candidate tag is calculated by using a second calculation formula, and the second calculation formula is set to Equation 2:

   

   In Equation 2, P _a,k represents the correlation of the target user u _a with the candidate tag t _k , U _K represents the user set using the tag t _k , U _N represents all users of the nearest neighbor set, Sa _{, u} Representing the similarity between the target user u _a and the specified user u _u , r _u, represents the correlation between the specified user u _u and the candidate tag t _k .
4. The electronic device according to claim 1, wherein said recommending an item related to the original label and the newly added label to the target user comprises:

   Selecting a label from the original label and the newly added label of the target user to obtain a set of items marked by the selected label, and calculating a correlation between the selected label and each item in the item set, in descending order of relevance. , selecting a third predetermined number of items to recommend to the target user.
5. The electronic device according to claim 4, wherein the correlation between the selection label and each item in the item set is calculated by using a third calculation formula, and the third calculation formula is set to formula 3:

   

   In Equation 3, t ₁ represents an original tag of the target user and a tag in the newly added tag set T _N , i represents an item in the item set I _t1 marked by the tag t ₁ , and relat(t ₁ , i) represents The correlation of the tag t ₁ with the item i, countUser(t ₁ , i) represents the number of users associated with item i and tag t ₁ , and countUser(t ₁ , j) represents the number of users associated with item j and tag t ₁ .
6. A collaborative filtering recommendation method is applied to an electronic device, and the method includes:

   Calculating the similarity between the target user and the different designated users according to the user-tag correlation matrix, and selecting the first predetermined number of designated users as the nearest neighbor set of the target user according to the order of similarity from high to low;

   Selecting, from the user tags of the nearest neighbor set, a tag that is not used by the target user as a candidate tag of the target user;

   Calculating a correlation between the target user and each candidate tag, and selecting a second predetermined number of candidate tags as the new tag of the target user according to the order of relevance from highest to lowest; and

   According to the original label and the newly added label of the target user, items related to the original label and the newly added label are recommended to the target user.
7. The collaborative filtering recommendation method according to claim 6, wherein the user-tag correlation matrix adopts a two-dimensional matrix, each row of the two-dimensional matrix represents a user, and each column represents a label, and the two-dimensional matrix Stores the correlation between all users and all tags;

   The similarity between the target user and different specified users is calculated by using a first calculation formula, and the first calculation formula is set to formula 1:

   

   In Equation 1, Sa _{a, s} represents the similarity between the target user u _a and the specified user u _s , and T _{a, s} represents the label used by the target user u _a and the specified user u _s , and r _{a, t} represents the target The correlation between the user u _a and the tag t, r _s,t represents the correlation between the specified user u _s and the tag t,

   

   Means the average of the correlation of user u _a with all tags,

   

   Represents the average of the correlation of user u _s with all tags.
8. The collaborative filtering recommendation method according to claim 6, wherein the correlation between the target user and each candidate tag is calculated by using a second calculation formula, and the second calculation formula is set to formula 2:

   

   In Equation 2, P _a,k represents the correlation of the target user u _a with the candidate tag t _k , U _K represents the user set using the tag t _k , U _N represents all users of the nearest neighbor set, Sa _{, u} Representing the similarity between the target user u _a and the specified user u _u , r _u, represents the correlation between the specified user u _u and the candidate tag t _k .
9. The collaborative filtering recommendation method according to claim 6, wherein the recommending the item related to the original label and the newly added label to the target user comprises:

   Selecting a label from the original label and the newly added label of the target user to obtain a set of items marked by the selected label, and calculating a correlation between the selected label and each item in the item set, in descending order of relevance. , selecting a third predetermined number of items to recommend to the target user.
10. The collaborative filtering recommendation method according to claim 9, wherein the correlation between the selection label and each item in the item set is calculated by using a third calculation formula, and the third calculation formula is set to formula 3:

    

    In Equation 3, t ₁ represents an original tag of the target user and a tag in the newly added tag set T _N , i represents an item in the item set I _t1 marked by the tag t ₁ , and relat(t ₁ , i) represents The correlation of the tag t ₁ with the item i, countUser(t ₁ , i) represents the number of users associated with item i and tag t ₁ , and countUser(t ₁ , j) represents the number of users associated with item j and tag t ₁ .
11. A computer readable storage medium, wherein the computer readable storage medium stores a collaborative filtering recommendation system, the collaborative filtering recommendation system being executable by at least one processor to cause the at least one processor to execute as follows step:

    Calculating the similarity between the target user and the different designated users according to the user-tag correlation matrix, and selecting the first predetermined number of designated users as the nearest neighbor set of the target user according to the order of similarity from high to low;

    Selecting, from the user tags of the nearest neighbor set, a tag that is not used by the target user as a candidate tag of the target user;

    Calculating a correlation between the target user and each candidate tag, and selecting a second predetermined number of candidate tags as the new tag of the target user according to the order of relevance from highest to lowest; and

    According to the original label and the newly added label of the target user, items related to the original label and the newly added label are recommended to the target user.
12. A computer readable storage medium according to claim 11, wherein said user-tag correlation matrix employs a two-dimensional matrix, each row of the two-dimensional matrix representing a user, each column representing a label, the two-dimensional The matrix stores the correlation between all users and all tags;

    The similarity between the target user and different specified users is calculated by using a first calculation formula, and the first calculation formula is set to formula 1:

    

    In Equation 1, Sa _{a, s} represents the similarity between the target user u _a and the specified user u _s , and T _{a, s} represents the label used by the target user u _a and the specified user u _s , and r _{a, t} represents the target The correlation between the user u _a and the tag t, r _s,t represents the correlation between the specified user u _s and the tag t,

    

    The average of the correlation between the user u _a and all tags,

    

    Represents the average of the correlation of user u _s with all tags.
13. The computer readable storage medium according to claim 11, wherein the correlation of the target user with each candidate tag is calculated using a second calculation formula, and the second calculation formula is set to Equation 2:

    

    In Equation 2, P _a,k represents the correlation of the target user u _a with the candidate tag t _k , U _K represents the user set using the tag t _k , U _N represents all users of the nearest neighbor set, Sa _{, u} Representing the similarity between the target user u _a and the specified user u _u , r _u, represents the correlation between the specified user u _u and the candidate tag t _k .
14. The computer readable storage medium of claim 11, wherein the recommending an item related to the original label and the newly added label to the target user comprises:

    Selecting a label from the original label and the newly added label of the target user to obtain a set of items marked by the selected label, and calculating a correlation between the selected label and each item in the item set, in descending order of relevance. , selecting a third predetermined number of items to recommend to the target user.
15. The computer readable storage medium according to claim 14, wherein the correlation between the selection label and each item in the item set is calculated by using a third calculation formula, and the third calculation formula is set to formula 3:

    

    In Equation 3, t ₁ represents an original label of the target user and a label in the newly added label set T _N , i represents an item in the item set I _t1 marked by the label t ₁ , and relat(t ₁ , i) represents The correlation of the tag t ₁ with the item i, countUser(t ₁ , i) represents the number of users associated with item i and tag t ₁ , and countUser(t ₁ , j) represents the number of users associated with item j and tag t ₁ .
16. A collaborative filtering recommendation system, comprising:

    a calculation module, configured to calculate a similarity between the target user and different specified users according to the user-tag correlation matrix, and select a first predetermined number of designated users as the nearest neighbor set of the target user according to the order of similarity from high to low;

    a selection module, configured to select, from the user tags of the nearest neighbor set, a tag that is not used by the target user, as a candidate tag of the target user;

    The calculating module is further configured to calculate a correlation between the target user and each candidate tag, and select a second predetermined number of candidate tags as a new tag of the target user according to a high to low correlation order;

    A recommendation module is configured to recommend an item related to the original label and the newly added label to the target user according to the original label and the newly added label of the target user.
17. The collaborative filtering recommendation system according to claim 16, wherein the user-tag correlation matrix adopts a two-dimensional matrix, each row of the two-dimensional matrix represents a user, and each column represents a label, and the two-dimensional matrix Stores the correlation between all users and all tags;

    The similarity between the target user and different specified users is calculated by using a first calculation formula, and the first calculation formula is set to formula 1:

    In Equation 1, Sa _{a, s} represents the similarity between the target user u _a and the specified user u _s , and T _{a, s} represents the label used by the target user u _a and the specified user u _s , and r _{a, t} represents the target The correlation between the user u _a and the tag t, r _s,t represents the correlation between the specified user u _s and the tag t,

    

    The average of the correlation between the user u _a and all tags,

    

    Represents the average of the correlation of user u _s with all tags.
18. The collaborative filtering recommendation system according to claim 16, wherein the correlation between the target user and each candidate tag is calculated by using a second calculation formula, and the second calculation formula is set to formula 2:

    

    In Equation 2, P _a,k represents the correlation of the target user u _a with the candidate tag t _k , U _K represents the user set using the tag t _k , U _N represents all users of the nearest neighbor set, Sa _{, u} Representing the similarity between the target user u _a and the specified user u _u , r _u, represents the correlation between the specified user u _u and the candidate tag t _k .
19. The collaborative filtering recommendation system according to claim 16, wherein the recommending the item related to the original label and the newly added label to the target user comprises:

    Selecting a label from the original label and the newly added label of the target user to obtain a set of items marked by the selected label, and calculating a correlation between the selected label and each item in the item set, in descending order of relevance. , selecting a third predetermined number of items to recommend to the target user.
20. The collaborative filtering recommendation system according to claim 19, wherein the correlation between the selection label and each item in the project set is calculated by using a third calculation formula, and the third calculation formula is set to Equation 3:

    

    In Equation 3, t ₁ represents an original tag of the target user and a tag in the newly added tag set T _N , i represents an item in the item set I _t1 marked by the tag t ₁ , and relat(t ₁ , i) represents The correlation of the tag t ₁ with the item i, countUser(t ₁ , i) represents the number of users associated with item i and tag t ₁ , and countUser(t ₁ , j) represents the number of users associated with item j and tag t ₁ .

Images