Search

CN-121980073-A - Resource recommendation method, device, equipment, storage medium and program product

CN121980073ACN 121980073 ACN121980073 ACN 121980073ACN-121980073-A

Abstract

The present disclosure provides a resource recommendation method, apparatus, device, storage medium, and program product, which can be applied to the big data field. The resource recommendation method comprises the steps of determining a second object matched with the preference of a target object from a plurality of candidate objects based on the interesting resources of the first object and the interesting resources of the plurality of candidate objects, extracting candidate resources different from the interesting resources of the first object from the interesting resources of the second object, analyzing historical interaction information of the second object aiming at the candidate resources and the similarity between the first object and the second object, generating expected interaction probability of the first object on the candidate resources, determining the target resources for recommending the first object from the candidate resources based on the expected interaction probability, and sending recommendation information of the target resources to the first object.

Inventors

  • Bei Fei
  • YANG YUXIN

Assignees

  • 中国工商银行股份有限公司

Dates

Publication Date
20260505
Application Date
20250918

Claims (12)

  1. 1. A method for recommending resources, the method comprising: Determining a second object from the plurality of candidate objects that matches the preference of the target object based on the resource of interest of the first object and the resource of interest of the plurality of candidate objects; extracting candidate resources different from the interesting resources of the first object from the interesting resources of the second object; Analyzing historical interaction information of the second object aiming at the candidate resource and the similarity between the first object and the second object to generate expected interaction probability of the first object on the candidate resource; Determining a target resource from the candidate resources for recommendation to the first object based on the expected interaction probability, and And sending the recommendation information of the target resource to the first object.
  2. 2. The method of claim 1, wherein the historical interaction information comprises extracting interaction operation transformation information and interaction result information for the candidate resource from an interaction process of selecting the candidate resource from at least two historical resources in a historical period; The analyzing the historical interaction information of the second object, the similarity between the first object and the second object and the candidate resource, and generating the expected interaction probability of the first object to the candidate resource includes: determining a degree of interest of the second object in the candidate resource based on the interactive operation transformation information and the interactive result information, and And generating the expected interaction probability based on the interest degree, the interaction result information and the interaction operation conversion information.
  3. 3. The method of claim 2, wherein the generating the expected interaction probability based on the degree of interest, the interaction result information, and the interaction transformation information comprises: determining a first interaction weight based on the interaction result information and the interaction operation conversion information; and generating the expected interaction probability based on the interest degree, the first interaction weight and the interaction result information.
  4. 4. The method of claim 3, wherein the generating the expected interaction probability based on the degree of interest, the interaction result information, and the interaction transformation information comprises: dynamically generating weight parameters matched with the candidate resources based on the interaction result information and the interaction operation conversion information by using the trained model; generating a second interaction weight based on the weight parameter, the interaction result information and the interaction operation conversion information; Generating the expected interaction probability based on the interest degree, the second interaction weight and the interaction result information; the trained model is obtained by training a label by taking an actual interaction result of a sample object on a sample resource.
  5. 5. The method of claim 1, wherein the determining a second object from the plurality of candidate objects that matches the preference of the target object based on the resource of interest of the first object and the resource of interest of the plurality of candidate objects comprises: generating a similarity between a first object and each candidate object based on the resource of interest of the first object and the resource of interest of the plurality of candidate objects, and Based on the similarity, a second object is determined from the plurality of candidate objects that matches the preference of the target object.
  6. 6. The method of claim 5, wherein generating the similarity between the first object and each candidate object based on the resource of interest of the first object and the resources of interest of the plurality of candidate objects comprises: determining target candidate resources of the same type as the resource of interest of the first object from the resources of interest of a plurality of candidate objects; Performing preference analysis on historical interaction information of the plurality of candidate objects aiming at the target candidate resource to generate preference information of the plurality of candidate objects; Performing preference analysis on the historical interaction information of the first object aiming at the target candidate resource to generate preference information of the first object, and And generating similarity between the first object and each candidate object based on the preference information of the first object and the preference information of the plurality of candidate objects.
  7. 7. The method of claim 1, wherein the determining a target resource from the candidate resources for recommendation to the first object based on the expected interaction probability comprises: Performing association analysis on the candidate resources to generate association degree among the candidate resources; determining a first resource from the candidate resources based on the expected interaction probability; Determining a second resource associated with the first resource from the candidate resources based on the degree of association, and And determining the first resource and the second resource as the target resource.
  8. 8. The method of claim 7, wherein the performing association analysis on the candidate resources to generate the association degree between the candidate resources comprises: Acquiring historical interaction information of the second object on the candidate resource; And carrying out association analysis on the candidate resources based on the historical interaction information to generate association degree among the candidate resources.
  9. 9. A resource recommendation device, the device comprising: A first determining module for determining a second object from the plurality of candidate objects that matches the preference of the target object based on the resource of interest of the first object and the resource of interest of the plurality of candidate objects; An extraction module for extracting candidate resources different from the resource of interest of the first object from the resource of interest of the second object; the analysis module is used for analyzing the historical interaction information of the second object aiming at the candidate resource and the similarity between the first object and the second object, and generating the expected interaction probability of the first object on the candidate resource; a second determination module for determining a target resource for recommending to the first object from the candidate resources based on the expected interaction probability, and And the recommending module is used for sending the recommending information of the target resource to the first object.
  10. 10. An electronic device, comprising: One or more processors; a memory for storing one or more computer programs, Characterized in that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1-8.
  11. 11. A computer-readable storage medium, on which a computer program or instructions is stored, which, when executed by a processor, carries out the steps of the method according to any one of claims 1-8.
  12. 12. A computer program product comprising a computer program or instructions which, when executed by a processor, implement the steps of the method according to any one of claims 1 to 8.

Description

Resource recommendation method, device, equipment, storage medium and program product Technical Field The present disclosure relates to the field of big data, and more particularly, to a resource recommendation method, apparatus, device, medium, and program product. Background On the e-commerce platform, the number of goods on the shelf per day can reach millions, and on the video platform, the video resources on the line per day are massive, so that the resources or contents of interest to the users are required to be accurately presented to the users. However, in the related art, the existing recommendation algorithm has low matching degree of the recommended resources for the target user, so that the target user has low interest degree on the recommended resources, and the resource recommendation accuracy is poor. Disclosure of Invention In view of the foregoing, the present disclosure provides resource recommendation methods, apparatuses, devices, media, and program products. According to a first aspect of the present disclosure, there is provided a resource recommendation method including determining a second object matching a preference of a target object from among a plurality of candidate objects based on a resource of interest of a first object and a resource of interest of the plurality of candidate objects, extracting a candidate resource different from the resource of interest of the first object from the resource of interest of the second object, analyzing historical interaction information of the second object with respect to the candidate resource, similarity between the first object and the second object, generating an expected interaction probability of the first object with respect to the candidate resource, determining the target resource for recommendation to the first object from the candidate resources based on the expected interaction probability, and transmitting recommendation information of the target resource to the first object. According to the embodiment of the disclosure, the historical interaction information comprises interaction operation conversion information and interaction result information aiming at candidate resources, which are extracted in the interaction process of selecting the candidate resources from at least two historical resources in a historical period, the historical interaction information of a second object, the similarity between the first object and the second object and the candidate resources are analyzed, the expected interaction probability of the first object on the candidate resources is generated, the expected interaction probability of the second object on the candidate resources is determined based on the interaction operation conversion information and the interaction result information, and the expected interaction probability is generated based on the interaction operation conversion information, the interaction result information and the interaction operation conversion information. According to an embodiment of the disclosure, the expected interaction probability is generated based on the degree of interest, the interaction result information and the interaction operation transformation information, and the method comprises the steps of determining a first interaction weight based on the interaction result information and the interaction operation transformation information, and generating the expected interaction probability based on the degree of interest, the first interaction weight and the interaction result information. According to the embodiment of the disclosure, expected interaction probability is generated based on the degree of interest, interaction result information and interaction operation conversion information, and the expected interaction probability is generated based on the interaction result information and the interaction operation conversion information by utilizing a trained model, wherein the trained model is obtained by training a tag by taking an actual interaction result of a sample object on the sample resource. According to an embodiment of the present disclosure, determining a second object from a plurality of candidate objects that matches a preference of a target object based on a resource of interest of a first object and a resource of interest of the plurality of candidate objects includes generating a similarity between the first object and each candidate object based on the resource of interest of the first object and the resource of interest of the plurality of candidate objects, and determining the second object from the plurality of candidate objects that matches the preference of the target object based on the similarity. According to the embodiment of the disclosure, the similarity between the first object and each candidate object is generated based on the interested resources of the first object and the interested resources of the plurality of candidate objects, and the method comprises the ste