Search

CN-121998047-A - Content recommendation system

CN121998047ACN 121998047 ACN121998047 ACN 121998047ACN-121998047-A

Abstract

The application provides a content recommendation system. The content recommendation system comprises the steps of obtaining interaction information to be processed, processing the interaction information to be processed according to target scene information to obtain target embedded information, and determining target recommendation information based on the target embedded information.

Inventors

  • FAN JIAQI

Assignees

  • 深圳市TCL高新技术开发有限公司

Dates

Publication Date
20260508
Application Date
20241105

Claims (13)

  1. 1. A method, wherein the content recommendation method comprises: acquiring interaction information to be processed; Processing the interaction information to be processed according to the target scene information to obtain target embedded information; And determining target recommendation information based on the target embedded information.
  2. 2. The method according to claim 1, wherein the processing the interaction information to be processed according to the target scene information to obtain target embedded information includes: determining popularity of the interactive items in the interactive information to be processed; Generating first embedded information corresponding to the interaction information to be processed; generating scene mask information of the interaction information to be processed according to the target scene information corresponding to the interaction information to be processed; and determining target embedded information according to the popularity of the interactive item, the first embedded information and the scene mask information.
  3. 3. The method according to claim 2, wherein generating scene mask information of the interaction information to be processed according to the target scene information corresponding to the interaction information to be processed includes: Acquiring interactive item information in the interactive information to be processed and scene identification information corresponding to the interactive item information; And determining target scene information corresponding to the interactive item information based on the scene identification information, and generating scene mask information corresponding to the target scene information.
  4. 4. The method of claim 3, wherein the determining, based on the scene identification information, target scene information corresponding to the interactive item information, and generating scene mask information corresponding to the target scene information, comprises: accessing a scene database, and acquiring candidate scene information associated with the interactive item information and candidate identification information of the candidate scene information in the scene database; determining candidate scene information, of which the candidate identification information is identical to the scene identification information, as target scene information of the interactive item information; And generating scene mask information of the target scene information according to the target scene information and the candidate scene information.
  5. 5. The method of claim 4, wherein generating scene mask information for the target scene information from the target scene information and the candidate scene information comprises: Acquiring a first scene value of the candidate scene information and a second scene value of the target scene information; And carrying out bit-wise or operation on the first scene value and the second scene value to obtain scene mask information of the target scene information.
  6. 6. The method of claim 2, wherein the determining target embedded information from the interactive item popularity, the first embedded information, and the scene mask information comprises: calculating scene popularity of candidate recommendation information based on the scene mask information and the interactive item popularity; and generating target embedded information of the interaction information to be processed according to the popularity of the scene, the first embedded information and a target processing model.
  7. 7. The method of any of claims 1-6, wherein the determining target recommendation information based on the target embedded information comprises: Acquiring target user embedded information and each item embedded information in the target embedded information; calculating the embedding similarity information between the item embedding information and the target user embedding information; And determining target item embedded information in the item embedded information according to the embedded similarity information, and acquiring target recommended information corresponding to the target item embedded information.
  8. 8. The method of claim 6, wherein prior to generating the target embedded information of the interaction information to be processed based on the scene popularity, the first embedded information, and the target processing model, further comprising: Performing first comparison learning according to the first embedded information and the scene popularity to obtain first loss information; performing second contrast learning according to the first embedded information and the scene mask information to obtain second loss information; Generating target loss information according to the main task loss information, the first loss information, the second loss information and the target super-parameters; and carrying out parameter adjustment on the current network model according to the target loss information to obtain a target processing model.
  9. 9. The method of claim 8, wherein the performing a first contrast study based on the first embedded information and the popularity of the scene to obtain first loss information comprises: Performing first separation operation on the first embedded information to obtain interest embedded information in the first embedded information; weighting the scene popularity by using a first popularity weight to obtain a first weighted popularity; And performing first comparison learning according to the interest embedded information and the weighted popularity to obtain first loss information.
  10. 10. The method of claim 8, wherein performing a second contrast learning based on the first embedded information and the scene mask information to obtain second loss information comprises: performing popular embedding separation on the first embedded information to obtain popular embedded information in the first embedded information; Weighting the scene popularity by using a second popularity weight to obtain a second weighted popularity; And performing second comparison learning according to the scene mask information, the popularity embedded information and the second weighted popularity to obtain second loss information.
  11. 11. A system, the system comprising: the information acquisition module is configured to acquire interaction information to be processed; The embedding extraction module is configured to process the interaction information to be processed according to the target scene information to obtain target embedding information; a content recommendation module configured to determine target recommendation information based on the target embedded information; Further, the processing the interaction information to be processed according to the target scene information to obtain target embedded information includes: determining popularity of the interactive items in the interactive information to be processed; Generating first embedded information corresponding to the interaction information to be processed; generating scene mask information of the interaction information to be processed according to the target scene information corresponding to the interaction information to be processed; Determining target embedded information according to the popularity of the interactive item, the first embedded information and the scene mask information; Further, the generating the scene mask information of the interaction information to be processed according to the target scene information corresponding to the interaction information to be processed includes: Acquiring interactive item information in the interactive information to be processed and scene identification information corresponding to the interactive item information; determining target scene information corresponding to the interactive item information based on the scene identification information, and generating scene mask information corresponding to the target scene information; Further, the determining, based on the scene identification information, the target scene information corresponding to the interactive item information, and generating the scene mask information corresponding to the target scene information, includes: accessing a scene database, and acquiring candidate scene information associated with the interactive item information and candidate identification information of the candidate scene information in the scene database; determining candidate scene information, of which the candidate identification information is identical to the scene identification information, as target scene information of the interactive item information; Generating scene mask information of the target scene information according to the target scene information and the candidate scene information; further, the generating scene mask information of the target scene information according to the target scene information and the candidate scene information includes: Acquiring a first scene value of the candidate scene information and a second scene value of the target scene information; performing bit-wise or operation on the first scene value and the second scene value to obtain scene mask information of the target scene information; further, the determining target embedded information according to the popularity of the interactive item, the first embedded information and the scene mask information includes: calculating scene popularity of candidate recommendation information based on the scene mask information and the interactive item popularity; generating target embedded information of the interaction information to be processed according to the popularity of the scene, the first embedded information and a target processing model; further, the determining target recommendation information based on the target embedded information includes: Acquiring target user embedded information and each item embedded information in the target embedded information; calculating the embedding similarity information between the item embedding information and the target user embedding information; Determining target item embedded information in the item embedded information according to the embedded similarity information, and acquiring target recommended information corresponding to the target item embedded information; further, before the generating the target embedded information of the interaction information to be processed according to the popularity of the scene, the first embedded information and the target processing model, the method further includes: Performing first comparison learning according to the first embedded information and the scene popularity to obtain first loss information; performing second contrast learning according to the first embedded information and the scene mask information to obtain second loss information; Generating target loss information according to the main task loss information, the first loss information, the second loss information and the target super-parameters; performing parameter adjustment on the current network model according to the target loss information to obtain a target processing model; Further, the performing a first comparison learning according to the first embedded information and the popularity of the scene to obtain first loss information includes: Performing first separation operation on the first embedded information to obtain interest embedded information in the first embedded information; weighting the scene popularity by using a first popularity weight to obtain a first weighted popularity; Performing first comparison learning according to the interest embedded information and the weighted popularity to obtain first loss information; Further, the performing second contrast learning according to the first embedded information and the scene mask information to obtain second loss information includes: performing popular embedding separation on the first embedded information to obtain popular embedded information in the first embedded information; Weighting the scene popularity by using a second popularity weight to obtain a second weighted popularity; And performing second comparison learning according to the scene mask information, the popularity embedded information and the second weighted popularity to obtain second loss information.
  12. 12. An apparatus, characterized in that the content recommendation apparatus comprises: One or more processors; Memory, and One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the processor to implement the steps of the method of any of claims 1 to 10.
  13. 13. A computer readable storage medium, having stored thereon a computer program, the computer program being loaded by a processor to perform the steps of the method of any of claims 1 to 10.

Description

Content recommendation system Technical Field The application relates to the technical field of computers, in particular to a content recommendation system. Background At present, with the rapid development of the internet field and various application programs, a recommendation system is provided in more and more application programs, the recommendation system is used for automatically calculating user preferences and providing high-quality recommendation services, the recommendation system is widely applied to various content recommendation, such as a network shopping platform, a music recommendation system, a movie recommendation system, a video-on-demand system and the like, and various contents such as graphics, pictures, videos and the like. The recommendation system can recommend the content of interest to the user through the user input information and the user history browsing record. Disclosure of Invention The embodiment of the application provides a content recommendation system. In one aspect, embodiments of the present application provide a method comprising the steps of: acquiring interaction information to be processed; Processing the interaction information to be processed according to the target scene information to obtain target embedded information; And determining target recommendation information based on the target embedded information. In another aspect, the present application provides a system comprising: the information acquisition module is configured to acquire interaction information to be processed; The embedding extraction module is configured to process the interaction information to be processed according to the target scene information to obtain target embedding information; and the content recommendation module is configured to determine target recommendation information based on the target embedded information. On the other hand, the present application also provides a content recommendation device, including: One or more processors; Memory, and One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the processor to implement the steps of the content recommendation method. In another aspect, the present application also provides a computer readable storage medium having stored thereon a computer program, the computer program being loaded by a processor to perform the steps of the content recommendation method. Drawings In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. FIG. 1 is a schematic view of a content recommendation method according to an embodiment of the present application; FIG. 2 is a flowchart of an embodiment of a content recommendation method according to an embodiment of the present application; FIG. 3 is a flowchart illustrating an embodiment of determining target recommendation information in a content recommendation method according to an embodiment of the present application; FIG. 4 is a flowchart of an embodiment of training a target processing model in a content recommendation method according to an embodiment of the present application; FIG. 5 is a schematic diagram illustrating a structure of an embodiment of a content recommendation system according to an embodiment of the present application; fig. 6 is a schematic structural diagram of an embodiment of a content recommendation device according to an embodiment of the present application. Detailed Description The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to fall within the scope of the application. In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. indicate orientations or positional relationships based on the drawings are merely for convenience in describing the present invention and simplifying the description, and do not indicate or imply that the apparatus or elements referred to must have a specific orientation, be configured and operated in a specific orientation, and thus should not be construed as limiting the present in