Product recommendation method and device, equipment, medium and product

Abstract

The invention provides a product recommendation method and device, equipment, a medium and a product, which not only considers the comprehensive evaluation (first crowd sensing score) of the product in a social group, but also integrates the family relationship comprehensive evaluation (second crowd sensing score) of a target user and the use habit comprehensive evaluation (third crowd sensing score) of the user. And comprehensive capture of multi-dimensional and deep demands of the user is realized. By integrating the diversified score information, the user preference can be described more accurately, and the recommendation result is prevented from falling into local optimum. Besides, a multi-target algorithm is adopted as a recommendation basis, optimal balance can be sought among a plurality of recommendation targets, it is ensured that recommended products can meet direct requirements of users and can also meet potential preferences and family environments of the users, and therefore the accuracy of product recommendation and the satisfaction degree of the users are remarkably improved.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a product recommendation method, device, equipment, medium and product. Background Art

[0002] With the rapid development of the Internet, the variety of goods on e-commerce platforms is increasing. While users enjoy a convenient shopping experience, they are also facing the problem of information overload. Accurately finding products that meet personal needs among a large number of goods has become a major challenge for users. At present, most e-commerce platforms use content-based recommendations or collaborative filtering recommendation methods. Although these methods can improve user experience to a certain extent, they often ignore the rich non-consumption behavior data of users and are difficult to fully capture the complex and changing needs of users. As a result, the recommendation results often fall into local optimality and cannot accurately match the real preferences of users, resulting in poor accuracy of product recommendations. Summary of the invention

[0003] The present application provides a product recommendation method, device, equipment, medium and product, which can improve the accuracy of product recommendations to target users.

[0004] In a first aspect, an embodiment of the present application provides a product recommendation method, the method comprising:

[0005] Obtaining a plurality of first group intelligence perception scores respectively corresponding to a plurality of products, wherein the first group intelligence perception scores are used to indicate a comprehensive evaluation of the product in a social group;

[0006] Obtaining a second crowd intelligence perception score corresponding to the target user's family relationship, where the second crowd intelligence perception score is used to indicate a comprehensive evaluation of the target user's family relationship in the social group;

[0007] Obtaining a third crowd intelligence perception score corresponding to the usage habit of the target user, wherein the usage habit is used to indicate a fixed behavior pattern of the target user in using a product, and the third crowd intelligence perception score is used to indicate a comprehensive evaluation of the usage habit of the target user in a social group;

[0008] Determine, according to the plurality of first crowd intelligence perception scores, the second crowd intelligence perception scores, and the third crowd intelligence perception scores, a plurality of comprehensive scores corresponding to the target user for the plurality of products, respectively, one comprehensive score corresponding to one product;

[0009] Based on a multi-objective algorithm, at least one target product is recommended to the target user based on the multiple comprehensive scores, the multiple products include the at least one target product, and the multi-objective algorithm is used to achieve multiple recommendation goals.

[0010] In one embodiment, the obtaining of a plurality of first group intelligence perception scores corresponding to a plurality of products mentioned above includes:

[0011] Obtain the number of evaluators of a first product, scores given to the first product by multiple users, and the level of the first product, wherein the first product is any one of the multiple products;

[0012] Calculating a first crowd intelligence perception score corresponding to the first product according to a first formula based on the number of evaluators, the scores of the first product respectively given by the multiple users, and the level of the first product;

[0013] Among them, the first formula is:

[0014]

[0015] is the first crowd intelligence perception score corresponding to the first product, P i is the number of reviewers of the first product, S i is the rating of the first product by the i-th user, i is a positive integer, 1≤i≤N, L i is the grade of the first product.

[0016] In one embodiment, the step of obtaining the second crowd intelligence perception score corresponding to the family relationship of the target user includes:

[0017] Obtaining the number of people who have evaluated the target user's family relationships, scores of the target user's family relationships by multiple systems, and the level of the target user's family relationships;

[0018] The second crowd intelligence perception score is calculated according to the second formula based on the number of evaluators of the target user's family relationship, the scores of the target user's family relationship by multiple systems, and the level of the target user's family relationship;

[0019] Wherein, the second formula is:

[0020]

[0021] Score the second group intelligence perception, P m is the number of evaluators of the target user’s family relationship, S m is the score of the target user's family relationship by the mth system, m is a positive integer, 1≤m≤M, L m The level of the target user's family relationship.

[0022] In one embodiment, the usage habits mentioned above include usage scope, usage frequency and usage duration;

[0023] The obtaining of a third crowd intelligence perception score corresponding to the usage habit of the target user includes:

[0024] Obtaining the number of evaluators of the target user's usage habits, first scores of the target user's usage scope by multiple systems, second scores of the target user's usage frequency by the multiple systems, third scores of the target user's usage time by the multiple systems, and the level of the target user's usage habits;

[0025] The third crowd intelligence perception score is calculated according to a third formula based on the number of evaluators of the target user's usage habits, the first scores of the target user's usage scope by multiple systems, the second scores of the target user's usage frequency by the multiple systems, the third scores of the target user's usage time by the multiple systems, and the level of the target user's usage habits;

[0026] Wherein, the third formula is:

[0027]

[0028] Score the third group intelligence perception, P t is the number of people who evaluated the usage habits of the target user, is the first score of the usage scope of the target user by the t-th system, is the second score of the usage frequency of the target user by the t-th system, is the third score of the usage time of the target user by the t-th system, t is a positive integer, 1≤t≤T, L t The level of usage habits of the target user.

[0029] In one embodiment, the determining of the multiple comprehensive scores of the target user corresponding to the multiple products respectively according to the multiple first crowd intelligence perception scores, the second crowd intelligence perception scores, and the third crowd intelligence perception scores mentioned above includes:

[0030] According to the multiple first crowd intelligence perception scores, the second crowd intelligence perception scores, and the third crowd intelligence perception scores, a fourth formula is used to calculate a plurality of comprehensive scores corresponding to the multiple products respectively;

[0031] Wherein, the fourth formula is:

[0032]

[0033] is the comprehensive score corresponding to the i-th product, k ais the first weight corresponding to the first crowd intelligence perception score, k r is the second weight corresponding to the second group intelligence perception score, k h is a third weight corresponding to the third crowd intelligence perception score, is the normalized representation of the first group intelligence perception score corresponding to the i-th product, is the normalized representation of the second crowd-sensing score, is a normalized representation of the third crowd sensing score.

[0034] In one embodiment, the method of recommending at least one target product to the target user based on the multiple comprehensive scores based on the multi-objective algorithm mentioned above includes:

[0035] Generate an initial product recommendation list using a variable neighborhood search algorithm according to the multiple comprehensive scores and the multiple recommendation targets;

[0036] Using a genetic algorithm, cross-mutate the initial product recommendation list to obtain multiple first recommendation lists;

[0037] Using a fast non-dominated sorting algorithm, filtering the plurality of first recommendation lists to obtain a second recommendation list;

[0038] The second recommendation list is optimized using a taboo search algorithm to obtain a target recommendation list, where the target recommendation list includes at least one target product recommended to the target user.

[0039] In one embodiment, the above-mentioned using a taboo search algorithm to optimize the second recommendation list to obtain a target recommendation list includes:

[0040] Randomly selecting two parent products from the second recommendation list, and performing crossover and mutation operations on the two selected parent products to generate a candidate recommendation list;

[0041] Performing a neighborhood search on the candidate recommendation list to obtain a first optimal recommendation list in the domain space;

[0042] In the case where the first optimal recommendation list does not exist in the preset taboo table, if the path length of the first optimal recommendation list is less than the path length of the current local optimal recommendation list, then the path length of the current local optimal recommendation list is updated to the path length of the first recommendation list, and the current local optimal recommendation list is updated to the first optimal recommendation list;

[0043] In the case where the updated path length of the current local optimal recommendation list is less than the path length of the current global optimal recommendation list, the current global optimal recommendation list is updated to the updated current local optimal recommendation list, and the updated current local optimal recommendation list is added to the taboo list;

[0044] Calculating, according to the multiple recommendation targets, a particle fitness value of an updated global optimal recommendation list;

[0045] When the fitness value of the particle is within a preset threshold range, or the current number of iterations reaches a preset number, the updated global optimal recommendation list is determined as the target recommendation list.

[0046] In a second aspect, the present application provides a product recommendation device, the device comprising:

[0047] A first acquisition module, used to acquire a plurality of first group intelligence perception scores corresponding to a plurality of products, wherein the first group intelligence perception scores are used to indicate a comprehensive evaluation of the product in a social group;

[0048] A second acquisition module, used to acquire a second crowd intelligence perception score corresponding to the target user's family relationship, where the second crowd intelligence perception score is used to indicate a comprehensive evaluation of the target user's family relationship in the social group;

[0049] A third acquisition module is used to acquire a third crowd intelligence perception score corresponding to the usage habits of the target user, wherein the usage habits are used to indicate a fixed behavior pattern of the target user in using the product, and the third crowd intelligence perception score is used to indicate a comprehensive evaluation of the usage habits of the target user in the social group;

[0050] a determination module, configured to determine, based on the plurality of first crowd intelligence perception scores, the second crowd intelligence perception scores, and the third crowd intelligence perception scores, a plurality of comprehensive scores corresponding to the target user for the plurality of products, respectively, one comprehensive score corresponding to one product;

[0051] A recommendation module is used to recommend at least one target product to the target user based on the multiple comprehensive scores based on a multi-objective algorithm, wherein the multiple products include the at least one target product, and the multi-objective algorithm is used to achieve multiple recommendation objectives.

[0052] In a third aspect, an embodiment of the present application provides an electronic device, the electronic device comprising: a processor and a memory storing computer program instructions;

[0053] When the processor executes the computer program instructions, the product recommendation method in any one of the embodiments of the first aspect is implemented.

[0054] In a fourth aspect, an embodiment of the present application provides a computer storage medium having computer program instructions stored thereon, and when the computer program instructions are executed by a processor, a product recommendation method as in any one of the embodiments in the first aspect is implemented.

[0055] In a fifth aspect, an embodiment of the present application provides a computer program product. When the instructions in the computer program product are executed by a processor of an electronic device, the electronic device executes a product recommendation method as in any one of the embodiments in the first aspect above.

[0056] In a product recommendation method, device, equipment, medium and product provided in an embodiment of the present application, not only the comprehensive evaluation of the product in the social group (first group intelligence perception score) is considered, but also the comprehensive evaluation of the target user's family relationship (second group intelligence perception score) and the comprehensive evaluation of the user's usage habits (third group intelligence perception score) are integrated, so as to achieve a comprehensive capture of the multi-dimensional and deep-level needs of users. By integrating these diversified rating information, user preferences can be more accurately characterized to avoid the recommendation results falling into the local optimum. In addition, by using a multi-objective algorithm as the basis for recommendation, it is possible to seek the best balance among multiple recommendation goals, ensuring that the recommended products can meet the user's direct needs and meet their potential preferences and family environment, thereby significantly improving the accuracy of product recommendations and user satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solution of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0058] Figure 1 It is a flowchart of a product recommendation method provided by an embodiment of the present application;

[0059] Figure 2 is a structural schematic diagram of a product recommendation device provided in an embodiment of the present application;

[0060] Figure 3 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0061] In order to more clearly understand the above-mentioned objectives, features and advantages of the present disclosure, the scheme of the present disclosure will be further described below. It should be noted that the embodiments of the present disclosure and the features in the embodiments can be combined with each other without conflict.

[0062] In the following description, many specific details are set forth to facilitate a full understanding of the present disclosure, but the present disclosure may also be implemented in other ways different from those described herein; it is obvious that the embodiments in the specification are only part of the embodiments of the present disclosure, rather than all of the embodiments.

[0063] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the term "comprising" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprising a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0064] Based on the background technology, we can see that currently, most e-commerce platforms use content-based recommendation or collaborative filtering recommendation methods. Although these methods can improve user experience to a certain extent, they often ignore the rich non-consumption behavior data of users and find it difficult to fully capture the complex and changeable needs of users. As a result, the recommendation results often fall into local optimality and cannot accurately match the user's real preferences, resulting in poor accuracy of product recommendations.

[0065] In order to solve the problems existing in the prior art, the embodiments of the present application provide a product recommendation method, device, equipment, medium and product, which not only considers the comprehensive evaluation of the product in the social group (the first group intelligence perception score), but also incorporates the comprehensive evaluation of the target user's family relationship (the second group intelligence perception score) and the comprehensive evaluation of the user's usage habits (the third group intelligence perception score), thereby achieving a comprehensive capture of the user's multi-dimensional and deep-level needs. By integrating these diversified rating information, user preferences can be more accurately portrayed, avoiding the recommendation results from falling into the local optimum. In addition, by using a multi-objective algorithm as the basis for recommendation, it is possible to seek the best balance among multiple recommendation goals, ensuring that the recommended products can meet the user's direct needs and meet their potential preferences and family environment, thereby significantly improving the accuracy of product recommendations and user satisfaction.

[0066] The present application embodiment provides a product recommendation method, device, equipment, medium and product. The following first introduces the product recommendation method provided by the present application embodiment. Figure 1 As shown, the method specifically comprises the following steps:

[0067] S101: Acquire a plurality of first group intelligence perception scores corresponding to a plurality of products, where the first group intelligence perception scores are used to indicate a comprehensive evaluation of the product in a social group;

[0068] S102: Obtain a second crowd intelligence perception score corresponding to the target user's family relationship, where the second crowd intelligence perception score is used to indicate a comprehensive evaluation of the target user's family relationship in the social group;

[0069] S103: Obtaining a third crowd intelligence perception score corresponding to the usage habit of the target user, where the usage habit is used to indicate a fixed behavior pattern of the target user in using a product, and the third crowd intelligence perception score is used to indicate a comprehensive evaluation of the usage habit of the target user in a social group;

[0070] S104: Determine, according to the multiple first crowd intelligence perception scores, the second crowd intelligence perception scores, and the third crowd intelligence perception scores, a plurality of comprehensive scores corresponding to the target user for the multiple products, respectively, where one product corresponds to one comprehensive score;

[0071] S105: Based on a multi-objective algorithm, recommend at least one target product to the target user based on the multiple comprehensive scores, the multiple products including the at least one target product, and the multi-objective algorithm is used to achieve multiple recommendation objectives.

[0072] The embodiments of the present application provide a product recommendation method, device, equipment, medium and product, which not only considers the comprehensive evaluation of the product in the social group (the first group intelligence perception score), but also incorporates the comprehensive evaluation of the target user's family relationship (the second group intelligence perception score) and the comprehensive evaluation of the user's usage habits (the third group intelligence perception score), thereby achieving a comprehensive capture of the user's multi-dimensional and deep-level needs. By integrating these diversified rating information, user preferences can be more accurately portrayed to avoid the recommendation results falling into the local optimum. In addition, by using a multi-objective algorithm as the basis for recommendation, it is possible to seek the best balance between multiple recommendation goals, ensuring that the recommended products can meet the user's direct needs and meet their potential preferences and family environment, thereby significantly improving the accuracy of product recommendations and user satisfaction.

[0073] In S101, for example, the first group intelligent perception rating may include comments on some food review websites, which can provide relevant information of many restaurants. Registered users can rate the environment, taste, and service according to their own food experience and express their own opinions. Other users can filter the restaurant's evaluation information before dining, so as to choose a restaurant that they are satisfied with.

[0074] In one example, obtaining a plurality of first crowd intelligence perception scores corresponding to a plurality of products respectively includes:

[0075] Obtain the number of evaluators of a first product, scores given to the first product by multiple users, and the level of the first product, wherein the first product is any one of the multiple products;

[0076] Calculating a first crowd intelligence perception score corresponding to the first product according to a first formula based on the number of evaluators, the scores of the first product respectively given by the multiple users, and the level of the first product;

[0077] Among them, the first formula is:

[0078]

[0079] is the first crowd intelligence perception score corresponding to the first product, P i is the number of reviewers of the first product, S i is the rating of the first product by the i-th user, i is a positive integer, 1≤i≤N, L i is the grade of the first product.

[0080] In the above embodiment, a specific embodiment is used as an example. For a product, the user's comments reflect the popularity and characteristics of the product to a certain extent. The number of people who evaluated the product ai is recorded as Pi, the rating of the user ui for the product ai is recorded as Si, the level of the product is recorded as Li, and the crowd intelligence perception social rating of the product is defined as The specific definition is shown in the following formula:

[0081]

[0082] Then perform normalization, see the following formula:

[0083]

[0084] in It is the product's maximum crowd intelligence perception social score. It is the minimum crowd-sensing social score of the product.

[0085] In S102, a second crowd-sensing score corresponding to the target user's family relationship is obtained, and the second crowd-sensing score is used to indicate the comprehensive evaluation of the target user's family relationship in the social group. Taking a specific embodiment as an example, for family relationships, the number of family relationship evaluators of user hm is recorded as Pm, the score of the family relationship of user hm by the system yu is Sm, and the level of the family relationship is recorded as Lm, and the crowd-sensing social score of the family relationship HGm is obtained. The specific definition is shown in the following formula:

[0086]

[0087] In one example,

[0088] The step of obtaining the second crowd intelligence perception score corresponding to the target user's family relationship includes:

[0089] Obtaining the number of people who have evaluated the target user's family relationships, scores of the target user's family relationships by multiple systems, and the level of the target user's family relationships;

[0090] The second crowd intelligence perception score is calculated according to the second formula based on the number of evaluators of the target user's family relationship, the scores of the target user's family relationship by multiple systems, and the level of the target user's family relationship;

[0091] Wherein, the second formula is:

[0092]

[0093] The second group intelligence perception score, P m is the number of evaluators of the target user’s family relationship, S m is the score of the target user's family relationship by the mth system, m is a positive integer, 1≤m≤M, L m The level of the target user's family relationship.

[0094] In S103, the third crowd intelligence perception score corresponding to the target user's usage habits is obtained. The usage habits are used to indicate the fixed behavior pattern of the target user in using the product. The third crowd intelligence perception score is used to indicate the comprehensive evaluation of the target user's usage habits in the social group. Taking a specific embodiment as an example, for usage habits, the number of evaluations of the usage habit tr is recorded as Pt, and the system yu scores the usage range tr of the usage habit as Frequency of use score The usage time rating is L v represents the total rating level, Lt represents the rating star of usage habits, and the crowd intelligence perception social rating of usage habits is obtained. The specific definition is shown in the following formula:

[0095]

[0096] In one example, the usage habits include usage scope, usage frequency, and usage duration;

[0097] The obtaining of a third crowd intelligence perception score corresponding to the usage habit of the target user includes:

[0098] Obtaining the number of evaluators of the target user's usage habits, first scores of the target user's usage scope by multiple systems, second scores of the target user's usage frequency by the multiple systems, third scores of the target user's usage time by the multiple systems, and the level of the target user's usage habits;

[0099] The third crowd intelligence perception score is calculated according to a third formula based on the number of evaluators of the target user's usage habits, the first scores of the target user's usage scope by multiple systems, the second scores of the target user's usage frequency by the multiple systems, the third scores of the target user's usage time by the multiple systems, and the level of the target user's usage habits;

[0100] Wherein, the third formula is:

[0101]

[0102] Score the third group intelligence perception, P t is the number of people who evaluated the usage habits of the target user, is the first score of the usage scope of the target user by the t-th system, is the second score of the usage frequency of the target user by the t-th system, is the third score of the usage time of the target user by the t-th system, t is a positive integer, 1≤t≤T, L t The level of usage habits of the target user.

[0103] In S104, multiple comprehensive scores corresponding to the target user for the multiple products are determined based on the multiple first crowd intelligence perception scores, the second crowd intelligence perception scores, and the third crowd intelligence perception scores, where one product corresponds to one comprehensive score. Taking a specific embodiment as an example, the comprehensive score is a comprehensive score of product popularity, family relationship, usage habit score, and crowd intelligence perception comprehensive score. The specific definition is shown in the following formula:

[0104]

[0105] where k a , k r , k h They are the weights corresponding to product popularity, family relationships, and usage habits.

[0106] In one example, determining, according to the plurality of first crowd intelligence perception scores, the second crowd intelligence perception scores, and the third crowd intelligence perception scores, a plurality of comprehensive scores corresponding to the target user with respect to the plurality of products, respectively, includes:

[0107] According to the multiple first crowd intelligence perception scores, the second crowd intelligence perception scores, and the third crowd intelligence perception scores, a fourth formula is used to calculate a plurality of comprehensive scores corresponding to the multiple products respectively;

[0108] Wherein, the fourth formula is:

[0109]

[0110] is the comprehensive score corresponding to the i-th product, k a is the first weight corresponding to the first crowd intelligence perception score, k r is the second weight corresponding to the second group intelligence perception score, k h is a third weight corresponding to the third crowd intelligence perception score, is the normalized representation of the first group intelligence perception score corresponding to the i-th product, is the normalized representation of the second crowd-sensing score, is a normalized representation of the third crowd sensing score.

[0111] In S105, based on the multi-objective algorithm, at least one target product is recommended to the target user based on the multiple comprehensive scores, the multiple products include the at least one target product, and the multi-objective algorithm is used to achieve multiple recommendation objectives. In the multi-objective hybrid taboo search algorithm, a basic solution is first obtained by a variable neighborhood search algorithm, which can meet the requirements of multi-objective and multi-constraints, then the basic solution is optimized by a hybrid particle swarm genetic optimization algorithm, and then the Pareto optimal solution is obtained by a fast non-dominated sorting algorithm, and finally the solution selected by the user is optimized by a taboo search algorithm. Traditional genetic algorithms cannot directly solve the multi-objective recommendation problem proposed in this chapter. Since genetic algorithms have better global optimization functions, and taboo search algorithms have better neighborhood search functions, the basic idea of ​​the hybrid taboo search algorithm is to first obtain a local optimal solution and mark it, and then avoid the local optimal solution in subsequent iterations, thereby expanding the search range. By comparing each local optimal solution, the global optimal solution is obtained. Therefore, the use of a hybrid taboo search optimization algorithm to solve the recommendation optimization problem can improve the recommendation accuracy.

[0112] In one example, the recommending at least one target product to the target user based on the multiple comprehensive scores based on the multi-objective algorithm includes:

[0113] Generate an initial product recommendation list using a variable neighborhood search algorithm according to the multiple comprehensive scores and the multiple recommendation targets;

[0114] Using a genetic algorithm, cross-mutate the initial product recommendation list to obtain multiple first recommendation lists;

[0115] Using a fast non-dominated sorting algorithm, filtering the plurality of first recommendation lists to obtain a second recommendation list;

[0116] The second recommendation list is optimized using a taboo search algorithm to obtain a target recommendation list, where the target recommendation list includes at least one target product recommended to the target user.

[0117] The above embodiment is described below by way of example according to a specific embodiment:

[0118] The specific steps are as follows.

[0119] (1) Initialization, optimization of the initial population. The initial solution generated by the variable neighborhood search algorithm is optimized by the genetic algorithm, and then the fast non-dominated sorting algorithm is used to obtain the Pareto optimal solution as the initial solution. See steps 1-2 for details.

[0120] (2) Determine whether the algorithm iteration termination condition is met. If yes, continue with the following steps to perform crossover and mutation of the parent generation to obtain the offspring. If no, output the result. See steps 3-5 for details.

[0121] (3) Generate neighborhood solutions using the neighborhood function of the current solution and determine several candidate solutions from them. If the neighborhood solution is not in the taboo table, compare the neighborhood solution and the current solution to see which one is better.

[0122] (4) Compare the current iterative optimal solution with the initial optimal solution. See steps 14-16 for details.

[0123] (5) The optimal solution after comparison is set as the initial solution for the next round of iteration, and the taboo table is judged.

[0124] (6) Calculate the solution fitness and determine whether the algorithm termination condition is met; if yes, terminate the iteration and output the optimization result; otherwise, go to step 3. See steps 19-26 for details.

[0125] In one example, the step of optimizing the second recommendation list by using a taboo search algorithm to obtain a target recommendation list includes:

[0126] Randomly selecting two parent products from the second recommendation list, and performing crossover and mutation operations on the two selected parent products to generate a candidate recommendation list;

[0127] Performing a neighborhood search on the candidate recommendation list to obtain a first optimal recommendation list in the domain space;

[0128] In the case where the first optimal recommendation list does not exist in the preset taboo table, if the path length of the first optimal recommendation list is less than the path length of the current local optimal recommendation list, then the path length of the current local optimal recommendation list is updated to the path length of the first recommendation list, and the current local optimal recommendation list is updated to the first optimal recommendation list;

[0129] In the case where the updated path length of the current local optimal recommendation list is less than the path length of the current global optimal recommendation list, the current global optimal recommendation list is updated to the updated current local optimal recommendation list, and the updated current local optimal recommendation list is added to the taboo list;

[0130] Calculating, according to the multiple recommendation targets, a particle fitness value of an updated global optimal recommendation list;

[0131] When the fitness value of the particle is within a preset threshold range, or the current number of iterations reaches a preset number, the updated global optimal recommendation list is determined as the target recommendation list.

[0132] The above embodiment is described below by way of example according to a specific embodiment:

[0133] Algorithm: Hybrid taboo search for multi-objective product recommendation

[0134] Input: product R, target value obj, number of iterations Threshold, number of neighborhood iterations

[0135] NeighborThreshold, Attraction A, Constraint res

[0136] Output: Product R1

[0137] Step 1. Initialize individual extreme values ​​according to the current product R and target obj

[0138] Step 2. Find the global extreme value subscript gBestLo of the current product

[0139] Step 3. while iterator Thresholdd do

[0140] Step 4. Randomly select the parent tempR and cross it with anotherR

[0141] Step 5. Mutation of the uncrossed parent chromosome tempR and the global extreme value gBestR

[0142] Step 6. While Nghiterator <NeighborThreshold do

[0143] Step 7. Find the neighborhood solution of the offspring

[0144] Step 8. If the neighborhood solution is not in the taboo table then

[0145] Step 9. ifLocalPathLen>tempPathLenthen

[0146] Step 10. Assign the value of tempPathLen to LocalPathLen

[0147] Step 11. end if

[0148] Step 12. end if

[0149] Step 13.end while

[0150] Step 14. if LocalPathLen <bestPathLenthen

[0151] Step 15. Assign the value of LocalPath to bestPath

[0152] Step 16. end if

[0153] Step 17. Set the solution of LocalPath as the initial solution

[0154] Step 18. Add the solution of LocalPath to the taboo table

[0155] Step 19. Calculate the particle fitness value of the optimal solution bestpath by combining the constraint condition res and the target value obj

[0156] Step 20. Update particles

[0157] Step 21. Compare the newly generated product bestpath with the global extreme value index gBestLoc

[0158] Step 22. if gBestLoc>=0 then

[0159] Step 23. Get the optimal solution gBestR1

[0160] Step 24. end if

[0161] Step 25. end while

[0162] Step 26. Output the optimal solution R1

[0163] Figure 2A schematic diagram of the structure of a product recommendation device provided in another embodiment of the present application is shown. For ease of explanation, only the parts related to the embodiment of the present application are shown.

[0164] Reference Figure 2 , the product recommendation device may include:

[0165] A first acquisition module 201 is used to acquire a plurality of first group intelligence perception scores corresponding to a plurality of products, wherein the first group intelligence perception scores are used to indicate a comprehensive evaluation of the product in a social group;

[0166] A second acquisition module 202 is used to acquire a second crowd intelligence perception score corresponding to the target user's family relationship, where the second crowd intelligence perception score is used to indicate a comprehensive evaluation of the target user's family relationship in the social group;

[0167] A third acquisition module 203 is used to acquire a third crowd intelligence perception score corresponding to the usage habit of the target user, wherein the usage habit is used to indicate a fixed behavior pattern of the target user in using the product, and the third crowd intelligence perception score is used to indicate a comprehensive evaluation of the usage habit of the target user in the social group;

[0168] A determination module 204 is configured to determine, based on the plurality of first crowd intelligence perception scores, the second crowd intelligence perception scores, and the third crowd intelligence perception scores, a plurality of comprehensive scores corresponding to the target user for the plurality of products, wherein one comprehensive score corresponds to one product;

[0169] The recommendation module 205 is used to recommend at least one target product to the target user based on the multiple comprehensive scores based on a multi-objective algorithm, wherein the multiple products include the at least one target product, and the multi-objective algorithm is used to achieve multiple recommendation objectives.

[0170] In one embodiment, the first acquisition module may be specifically used for:

[0171] Obtain the number of evaluators of a first product, scores given to the first product by multiple users, and the level of the first product, wherein the first product is any one of the multiple products;

[0172] Calculating a first crowd intelligence perception score corresponding to the first product according to a first formula based on the number of evaluators, the scores of the first product respectively given by the multiple users, and the level of the first product;

[0173] Among them, the first formula is:

[0174]

[0175] is the first crowd intelligence perception score corresponding to the first product, P i is the number of reviewers of the first product, S i is the rating of the first product by the i-th user, i is a positive integer, 1≤i≤N, L i is the grade of the first product.

[0176] In one embodiment, the second acquisition module may be specifically used for:

[0177] Obtaining the number of people who have evaluated the target user's family relationships, scores of the target user's family relationships by multiple systems, and the level of the target user's family relationships;

[0178] The second crowd intelligence perception score is calculated according to the second formula based on the number of evaluators of the target user's family relationship, the scores of the target user's family relationship by multiple systems, and the level of the target user's family relationship;

[0179] Wherein, the second formula is:

[0180]

[0181] The second group intelligence perception score, P m is the number of evaluators of the target user’s family relationship, S m is the score of the target user's family relationship by the mth system, m is a positive integer, 1≤m≤M, L m The level of the target user's family relationship.

[0182] In one embodiment, the third acquisition module may be specifically used for:

[0183] The obtaining of a third crowd intelligence perception score corresponding to the usage habit of the target user includes:

[0184] Obtaining the number of evaluators of the target user's usage habits, first scores of the target user's usage scope by multiple systems, second scores of the target user's usage frequency by the multiple systems, third scores of the target user's usage time by the multiple systems, and the level of the target user's usage habits;

[0185] The third crowd intelligence perception score is calculated according to a third formula based on the number of evaluators of the target user's usage habits, the first scores of the target user's usage scope by multiple systems, the second scores of the target user's usage frequency by the multiple systems, the third scores of the target user's usage time by the multiple systems, and the level of the target user's usage habits;

[0186] Wherein, the third formula is:

[0187]

[0188] Score the third group intelligence perception, P t is the number of people who evaluated the usage habits of the target user, is the first score of the usage scope of the target user by the t-th system, is the second score of the usage frequency of the target user by the t-th system, is the third score of the usage time of the target user by the t-th system, t is a positive integer, 1≤t≤T, L t The level of usage habits of the target user.

[0189] In one embodiment, the determination module may be specifically used to:

[0190] According to the multiple first crowd intelligence perception scores, the second crowd intelligence perception scores, and the third crowd intelligence perception scores, a fourth formula is used to calculate a plurality of comprehensive scores corresponding to the multiple products respectively;

[0191] Wherein, the fourth formula is:

[0192]

[0193] is the comprehensive score corresponding to the i-th product, k a is the first weight corresponding to the first crowd intelligence perception score, k r is the second weight corresponding to the second group intelligence perception score, k h is a third weight corresponding to the third crowd intelligence perception score, is the normalized representation of the first group intelligence perception score corresponding to the i-th product, is the normalized representation of the second crowd-sensing score, is a normalized representation of the third crowd sensing score.

[0194] In one embodiment, the recommendation module may include:

[0195] A generating unit, configured to generate an initial product recommendation list using a variable neighborhood search algorithm according to the multiple comprehensive scores and the multiple recommendation targets;

[0196] A mutation unit, configured to perform crossover mutation on the initial product recommendation list using a genetic algorithm to obtain a plurality of first recommendation lists;

[0197] A screening unit, configured to screen the plurality of first recommendation lists to obtain a second recommendation list by using a fast non-dominated sorting algorithm;

[0198] The optimization unit is used to optimize the second recommendation list by using a taboo search algorithm to obtain a target recommendation list, wherein the target recommendation list includes recommending at least one target product to the target user.

[0199] In one embodiment, the optimization unit may be specifically configured to:

[0200] Randomly selecting two parent products from the second recommendation list, and performing crossover and mutation operations on the two selected parent products to generate a candidate recommendation list;

[0201] Performing a neighborhood search on the candidate recommendation list to obtain a first optimal recommendation list in the domain space;

[0202] In the case where the first optimal recommendation list does not exist in the preset taboo table, if the path length of the first optimal recommendation list is less than the path length of the current local optimal recommendation list, then the path length of the current local optimal recommendation list is updated to the path length of the first recommendation list, and the current local optimal recommendation list is updated to the first optimal recommendation list;

[0203] In the case where the updated path length of the current local optimal recommendation list is less than the path length of the current global optimal recommendation list, the current global optimal recommendation list is updated to the updated current local optimal recommendation list, and the updated current local optimal recommendation list is added to the taboo list;

[0204] Calculating, according to the multiple recommendation targets, a particle fitness value of an updated global optimal recommendation list;

[0205] When the fitness value of the particle is within a preset threshold range, or the current number of iterations reaches a preset number, the updated global optimal recommendation list is determined as the target recommendation list.

[0206] It should be noted that the information interaction, execution process, etc. between the above-mentioned devices / units are based on the same concept as the method embodiment of the present application, and are devices corresponding to the above-mentioned product recommendation method. All implementation methods in the above-mentioned method embodiment are applicable to the embodiment of the device. Its specific functions and technical effects brought about can be found in the method embodiment part, which will not be repeated here.

[0207] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0208] Figure 3 A schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present application is shown.

[0209] The device may include a processor 301 and a memory 302 storing program instructions.

[0210] When the processor 301 executes the program, the steps in any of the above method embodiments are implemented.

[0211] Exemplarily, the program may be divided into one or more modules / units, one or more modules / units are stored in the memory 302, and executed by the processor 301 to complete the present application. One or more modules / units may be a series of program instruction segments capable of completing a specific function, and the instruction segments are used to describe the execution process of the program in the device.

[0212] Specifically, the processor 301 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0213] The memory 302 may include a large capacity memory for data or instructions. By way of example and not limitation, the memory 302 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive or a combination of two or more of these. In appropriate cases, the memory 302 may include a removable or non-removable (or fixed) medium. In appropriate cases, the memory 302 may be inside or outside the integrated gateway disaster recovery device. In a specific embodiment, the memory 302 is a non-volatile solid-state memory.

[0214] The memory may include read-only memory (ROM), random access memory (RAM), magnetic disk storage media devices, optical storage media devices, flash memory devices, electrical, optical or other physical / tangible memory storage devices. Thus, typically, the memory includes one or more tangible (non-transitory) readable storage media (e.g., memory devices) encoded with software including computer executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to an aspect of the present disclosure.

[0215] The processor 301 implements any one of the methods in the above embodiments by reading and executing program instructions stored in the memory 302 .

[0216] In one example, the electronic device may further include a communication interface 303 and a bus 310. The processor 301, the memory 302, and the communication interface 303 are connected via the bus 310 and communicate with each other.

[0217] The communication interface 303 is mainly used to implement communication between various modules, devices, units and / or equipment in the embodiments of the present application.

[0218] Bus 310 includes hardware, software or both, and the parts of online data flow billing equipment are coupled to each other. For example, but not limitation, bus may include accelerated graphics port (AGP) or other graphics bus, enhanced industrial standard architecture (EISA) bus, front-end bus (FSB), hypertransport (HT) interconnection, industrial standard architecture (ISA) bus, infinite bandwidth interconnection, low pin count (LPC) bus, memory bus, micro channel architecture (MCA) bus, peripheral component interconnection (PCI) bus, PCI-Express (PCI-X) bus, serial advanced technology attachment (SATA) bus, video electronics standard association local (VLB) bus or other suitable bus or two or more of these combinations. In appropriate cases, bus 310 may include one or more buses. Although the present application embodiment describes and shows a specific bus, the present application considers any suitable bus or interconnection.

[0219] In addition, in combination with the method in the above embodiment, the embodiment of the present application can provide a storage medium for implementation. The storage medium stores program instructions; when the program instructions are executed by a processor, any one of the methods in the above embodiment is implemented.

[0220] An embodiment of the present application further provides a chip, which includes a processor and a communication interface, wherein the communication interface and the processor are coupled, and the processor is used to run programs or instructions to implement the various processes of the above-mentioned method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0221] It should be understood that the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.

[0222] An embodiment of the present application provides a computer program product, which is stored in a storage medium. The program product is executed by at least one processor to implement the various processes of the above-mentioned method embodiment and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0223] It should be clear that the present application is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present application is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present application.

[0224] The functional modules shown in the above block diagram can be implemented as hardware, software, firmware or their combination. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), suitable firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present application are programs or code segments that are used to perform the required tasks. Programs or code segments can be stored in machine-readable media, or transmitted on a transmission medium or a communication link by a data signal carried in a carrier wave. "Machine-readable media" can include any medium that can store or transmit information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, optical fiber media, radio frequency (RF) links, etc. Code segments can be downloaded via computer grids such as the Internet, intranets, etc.

[0225] It should also be noted that the exemplary embodiments mentioned in this application describe some methods or systems based on a series of steps or devices. However, this application is not limited to the order of the above steps, that is, the steps can be performed in the order mentioned in the embodiment, or in a different order from the embodiment, or several steps can be performed simultaneously.

[0226] Aspects of the present disclosure are described above with reference to the flowchart and / or block diagram of the method, device (system) and program product according to the embodiment of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer program instructions. These program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer or other programmable data processing device to produce a machine so that these instructions executed by the processor of the computer or other programmable data processing device enable the implementation of the function / action specified in one or more boxes of the flowchart and / or block diagram. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor or a field programmable logic circuit. It can also be understood that each box in the block diagram and / or flowchart and the combination of boxes in the block diagram and / or flowchart can also be implemented by dedicated hardware that performs a specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions.

[0227] The above are only specific implementation methods of the present application. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, modules and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. It should be understood that the protection scope of the present application is not limited to this. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in this application, and these modifications or replacements should be included in the protection scope of this application.

Claims

1. A product recommendation method, characterized in that: The method comprises: Obtaining a plurality of first group intelligence perception scores respectively corresponding to a plurality of products, wherein the first group intelligence perception scores are used to indicate a comprehensive evaluation of the product in a social group; Obtaining a second crowd intelligence perception score corresponding to the target user's family relationship, where the second crowd intelligence perception score is used to indicate a comprehensive evaluation of the target user's family relationship in the social group; Obtaining a third crowd intelligence perception score corresponding to the usage habit of the target user, wherein the usage habit is used to indicate a fixed behavior pattern of the target user in using a product, and the third crowd intelligence perception score is used to indicate a comprehensive evaluation of the usage habit of the target user in a social group; Determine, according to the plurality of first crowd intelligence perception scores, the second crowd intelligence perception scores, and the third crowd intelligence perception scores, a plurality of comprehensive scores corresponding to the target user for the plurality of products, respectively, one comprehensive score corresponding to one product; Based on a multi-objective algorithm, at least one target product is recommended to the target user based on the multiple comprehensive scores, the multiple products include the at least one target product, and the multi-objective algorithm is used to achieve multiple recommendation goals.

2. The method according to claim 1, characterized in that The obtaining of a plurality of first group intelligence perception scores respectively corresponding to a plurality of products includes: Obtain the number of evaluators of a first product, scores given to the first product by multiple users, and the level of the first product, wherein the first product is any one of the multiple products; Calculating a first crowd intelligence perception score corresponding to the first product according to a first formula based on the number of evaluators, the scores of the first product respectively given by the multiple users, and the level of the first product; Among them, the first formula is: is the first crowd intelligence perception score corresponding to the first product, P i is the number of reviewers of the first product, S i is the rating of the first product by the i-th user, i is a positive integer, 1≤i≤N, L i is the grade of the first product.

3. The method according to claim 1, characterized in that The step of obtaining the second crowd intelligence perception score corresponding to the target user's family relationship includes: Obtaining the number of people who have evaluated the target user's family relationships, scores of the target user's family relationships by multiple systems, and the level of the target user's family relationships; The second crowd intelligence perception score is calculated according to the second formula based on the number of evaluators of the target user's family relationship, the scores of the target user's family relationship by multiple systems, and the level of the target user's family relationship; Wherein, the second formula is: The second group intelligence perception score, P m is the number of evaluators of the target user’s family relationship, S m is the score of the target user's family relationship by the mth system, m is a positive integer, 1≤m≤M, L m The level of the target user's family relationship.

4. The method according to claim 1, characterized in that: The usage habits include usage scope, usage frequency and usage duration; The obtaining of a third crowd intelligence perception score corresponding to the usage habit of the target user includes: Obtaining the number of evaluators of the target user's usage habits, first scores of the target user's usage scope by multiple systems, second scores of the target user's usage frequency by the multiple systems, third scores of the target user's usage time by the multiple systems, and the level of the target user's usage habits; The third crowd intelligence perception score is calculated according to a third formula based on the number of evaluators of the target user's usage habits, the first scores of the target user's usage scope by multiple systems, the second scores of the target user's usage frequency by the multiple systems, the third scores of the target user's usage time by the multiple systems, and the level of the target user's usage habits; Wherein, the third formula is: Score the third group intelligence perception, P t is the number of people who evaluated the usage habits of the target user, is the first score of the usage scope of the target user by the t-th system, is the second score of the usage frequency of the target user by the t-th system, is the third score of the usage time of the target user by the t-th system, t is a positive integer, 1≤t≤T, L t The level of usage habits of the target user.

5. The method according to claim 1, characterized in that The determining, according to the plurality of first crowd intelligence perception scores, the second crowd intelligence perception scores, and the third crowd intelligence perception scores, a plurality of comprehensive scores corresponding to the target user with respect to the plurality of products, respectively, comprises: According to the multiple first crowd intelligence perception scores, the second crowd intelligence perception scores, and the third crowd intelligence perception scores, a fourth formula is used to calculate and obtain multiple comprehensive scores corresponding to the multiple products respectively; Wherein, the fourth formula is: is the comprehensive score corresponding to the i-th product, k a is the first weight corresponding to the first crowd intelligence perception score, k r is the second weight corresponding to the second group intelligence perception score, k h is a third weight corresponding to the third crowd intelligence perception score, is the normalized representation of the first group intelligence perception score corresponding to the i-th product, is the normalized representation of the second crowd-sensing score, is a normalized representation of the third crowd sensing score.

6. The method according to claim 1, characterized in that The method of recommending at least one target product to the target user based on the multiple comprehensive scores based on the multi-objective algorithm includes: Generate an initial product recommendation list using a variable neighborhood search algorithm according to the multiple comprehensive scores and the multiple recommendation targets; Using a genetic algorithm, cross-mutate the initial product recommendation list to obtain multiple first recommendation lists; Using a fast non-dominated sorting algorithm, filtering the plurality of first recommendation lists to obtain a second recommendation list; The second recommendation list is optimized using a taboo search algorithm to obtain a target recommendation list, where the target recommendation list includes at least one target product recommended to the target user.

7. The method according to claim 6, characterized in that The step of optimizing the second recommendation list by using a taboo search algorithm to obtain a target recommendation list includes: Randomly selecting two parent products from the second recommendation list, and performing crossover and mutation operations on the two selected parent products to generate a candidate recommendation list; Performing a neighborhood search on the candidate recommendation list to obtain a first optimal recommendation list in the domain space; In the case that the first optimal recommendation list does not exist in the preset taboo table, if the path length of the first optimal recommendation list is less than the path length of the current local optimal recommendation list, the path length of the current local optimal recommendation list is updated to the path length of the first recommendation list, and the current local optimal recommendation list is updated to the first optimal recommendation list; When the updated path length of the current local optimal recommendation list is less than the path length of the current global optimal recommendation list, the current global optimal recommendation list is updated to the updated current local optimal recommendation list, and the updated current local optimal recommendation list is added to the taboo list; Calculating, according to the multiple recommendation targets, a particle fitness value of an updated global optimal recommendation list; When the fitness value of the particle is within a preset threshold range, or the current number of iterations reaches a preset number, the updated global optimal recommendation list is determined as the target recommendation list.

8. A product recommendation device, characterized in that: The device comprises: A first acquisition module, used to acquire a plurality of first group intelligence perception scores corresponding to a plurality of products, wherein the first group intelligence perception scores are used to indicate a comprehensive evaluation of the product in a social group; A second acquisition module, used to acquire a second crowd intelligence perception score corresponding to the target user's family relationship, where the second crowd intelligence perception score is used to indicate a comprehensive evaluation of the target user's family relationship in the social group; A third acquisition module is used to acquire a third crowd intelligence perception score corresponding to the usage habits of the target user, wherein the usage habits are used to indicate a fixed behavior pattern of the target user in using the product, and the third crowd intelligence perception score is used to indicate a comprehensive evaluation of the usage habits of the target user in the social group; a determination module, configured to determine, based on the plurality of first crowd intelligence perception scores, the second crowd intelligence perception scores, and the third crowd intelligence perception scores, a plurality of comprehensive scores corresponding to the target user for the plurality of products, respectively, one comprehensive score corresponding to one product; A recommendation module is used to recommend at least one target product to the target user based on the multiple comprehensive scores based on a multi-objective algorithm, wherein the multiple products include the at least one target product, and the multi-objective algorithm is used to achieve multiple recommendation objectives.

9. An electronic device, characterized in that: The device comprises: a processor and a memory storing computer program instructions; When the processor executes the computer program instructions, the product recommendation method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer program instructions, and when the computer program instructions are executed by a processor, the product recommendation method according to any one of claims 1 to 7 is implemented.

11. A computer program product, characterized in that When the instructions in the computer program product are executed by a processor of an electronic device, the electronic device executes the product recommendation method as described in any one of claims 1 to 7.

Images