WO2023071404A1 - Procédé de recommandation, dispositif électronique et support de stockage - Google Patents

Procédé de recommandation, dispositif électronique et support de stockage Download PDF

Info

Publication number
WO2023071404A1
WO2023071404A1 PCT/CN2022/112106 CN2022112106W WO2023071404A1 WO 2023071404 A1 WO2023071404 A1 WO 2023071404A1 CN 2022112106 W CN2022112106 W CN 2022112106W WO 2023071404 A1 WO2023071404 A1 WO 2023071404A1
Authority
WO
WIPO (PCT)
Prior art keywords
user
service
information
historical behavior
server
Prior art date
Application number
PCT/CN2022/112106
Other languages
English (en)
Chinese (zh)
Inventor
刘辉
Original Assignee
花瓣云科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 花瓣云科技有限公司 filed Critical 花瓣云科技有限公司
Publication of WO2023071404A1 publication Critical patent/WO2023071404A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/953—Querying, e.g. by the use of web search engines
    • G06F16/9535—Search customisation based on user profiles and personalisation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
    • H04N21/25—Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies

Definitions

  • the present application relates to the field of data processing, and in particular to a recommendation method, electronic equipment and a storage medium.
  • Personalized recommendation generally refers to discovering the user's individual needs and interest characteristics by analyzing and mining the user's historical behavior information, and recommending information or products that the user may be interested in to the user.
  • Personalized recommendations can be applied to e-commerce systems, social networks, advertising recommendations, search engines, and many other aspects.
  • the video platform can also include a recommendation system dedicated to personalized recommendation.
  • the recommendation system can analyze and mine historical videos watched by users to find out which types of videos users prefer. more interesting, and recommend videos that are more interesting to users.
  • the recommendation system can also evaluate the similarity between the user and other users, and based on the user collaborative filtering (user collaboration filter, UserCF) algorithm, recommend to the user videos that are similar to the user and other users are interested in.
  • the movies that user A has watched many times are all sci-fi movies, such as: sci-fi movie 1, sci-fi movie 2, etc.
  • the recommendation system finds that user B has also watched sci-fi movie 1 and sci-fi movie 2 many times through data analysis, and User B has also watched science fiction movie 3 many times, then the recommendation system can recommend science fiction movie 3 to user A.
  • personalized recommendation often faces incremental problems, such as: personalized recommendation for new users.
  • new users do not have historical behavior information for the recommendation system of personalized recommendation, and the recommendation system cannot rely on the user's historical behavior information to make personalized recommendations to users.
  • the present application provides a recommendation method, an electronic device and a storage medium, which can reduce the results of user grouping, effectively improve the accuracy of personalized recommendation, and recommend more interesting recommendation results for users.
  • the present application provides a recommendation method, the method includes: the first device sends the user information of the first user to the server; the server sends recommendation information to the first device; the recommendation information is related to the first user group Information related to the user's use of the first service; the user information of the first user is related to the first user group; the first user group is based on the characteristic information corresponding to the historical behavior information related to the first user in the second service, and other The feature information corresponding to the historical behavior information of one or more users in the second service is determined by matching.
  • the method determines that the first user is in User grouping results in the second service, and based on the user grouping results of the first user in the second service, personalized recommendations are made to the first user in the first service, which can narrow down the results of user grouping for the first user, Effectively improve the accuracy of personalized recommendation, and recommend more interesting recommendation results for the first user.
  • the first device includes a first service and a second service; the method further includes: the first device sends feature information corresponding to historical behavior information related to the first user in the second service to the server.
  • both the first service and the second service may be provided by the first device.
  • the historical behavior information related to the first user in the second service may include: the corresponding historical behavior information of the first user in the second service, or other accounts (family public account) associated with the first user's account or other user's account) corresponding historical behavior information in the second service.
  • the first device includes a first service
  • the second device includes a second service
  • the method further includes: the second device sends historical behavior information related to the first user in the second service to the server Corresponding feature information.
  • the first service may be provided by the first device, and the second service may be provided by the second device.
  • the first device and the second device may be two terminal devices in a multi-device coordination scenario.
  • the sending, by the first device, the user information of the first user to the server includes: sending, by the first device, the user information of the first user to the server through a third device.
  • the server sending the recommendation information to the first device includes: the server sending the recommendation information to the first device through the third device.
  • the first device may be a device with less device resources and cannot be independently networked and/or human-computer interaction
  • the third device may be a device with relatively rich device resources and independent network and human-computer interaction.
  • the sending by the first device to the server of the characteristic information corresponding to the historical behavior information related to the first user in the second service may include: the first device sends to the server the feature information related to the first user in the second service through the third device Feature information corresponding to user-related historical behavior information.
  • the second device sending to the server the characteristic information corresponding to the historical behavior information related to the first user in the second service may include: the second device sends to the server through the fourth device Feature information corresponding to user-related historical behavior information.
  • the second device may be a device with less device resources, which cannot be independently networked and/or human-computer interaction
  • the fourth device may be a device with relatively rich device resources, which may be independently networked and human-computer interaction.
  • the user information of the first user includes: identification information of the first user, and/or identification information of the first user group.
  • the identification information of the first user may be a user account of the first user.
  • the method further includes: the server according to the characteristic information corresponding to the historical behavior information related to the first user in the second service and the characteristic information corresponding to the historical behavior information of one or more other users in the second service , determine the first user group; the server sends the identification information of the first user group to the first device.
  • the historical behavior information of other users in the first user group in the second service is similar to the historical behavior information related to the first user in the second service.
  • the characteristic information corresponding to the historical behavior information related to the first user in the second service includes: a hash value corresponding to the historical behavior information related to the first user in the second service.
  • the feature information corresponding to the historical behavior information of the other one or more users in the second service includes: a hash value corresponding to the historical behavior information of the other one or more users in the second service.
  • the hash value corresponding to the historical behavior information related to the first user in the second service is obtained by hashing the first encoding result using the first hash algorithm, and the first encoding result is obtained by using the first encoding
  • the method is obtained by encoding the historical behavior information related to the first user in the second service.
  • the hash value corresponding to the historical behavior information of one or more users in the second service is obtained by hashing the second encoding result using the first hash algorithm, and the second encoding result is obtained by using the first encoding method to It is obtained by encoding the historical behavior information of one or more other users in the second service.
  • the first hash algorithm may include: a locality sensitive hash (locality sensitive hashing, LSH) algorithm, a minimum hash (min-hash) algorithm using Jaccard (jaccard) to measure data similarity, a P-stable hash (P-stable hash) algorithm, simhash algorithm, etc., which measure data similarity by distance.
  • LSH locality sensitive hashing
  • min-hash minimum hash
  • Jaccard Jaccard
  • P-stable hash P-stable hash
  • simhash algorithm simhash algorithm
  • the first hash algorithm has the characteristic of maintaining the similarity of hash values after hash calculation is performed on the data to be hashed.
  • the first encoding manner may include: one-hot encoding, index (index) encoding (dictionary-based encoding), multi-hot (multiple-hot) encoding, and the like.
  • the encoding results (such as the first encoding result and the second encoding result) corresponding to different historical behavior information are unique.
  • the first device sending the user information of the first user to the server includes: when detecting that the first user uses the first service for the first time, the first device sends the first user's information to the server. User Info.
  • making personalized recommendations to the first user when the first user uses the first service for the first time can be referred to as performing a user cold start for the first user.
  • This method can improve the recommendation for the first user when performing a user cold start. Accuracy, the effect of user cold start can be better.
  • the present application provides a recommendation system, and the recommendation system can implement the recommendation method described in the first aspect and any possible implementation manner of the first aspect.
  • the recommendation system may include: a first device and a server.
  • the first device may be configured to send user information of the first user to the server.
  • the server can be used to send recommendation information to the first device.
  • the recommended information is information related to the use of the first service by users in the first user group; the user information of the first user is related to the first user group; the first user group is based on the information related to the first user in the second service.
  • the characteristic information corresponding to the historical behavior information of the user and the characteristic information corresponding to the historical behavior information of the other one or more users in the second service are determined by matching.
  • the first device includes a first service and a second service; the first device is further configured to send feature information corresponding to historical behavior information related to the first user in the second service to the server.
  • the recommendation system further includes a second device.
  • the first device includes a first service and the second device includes a second service.
  • the second device may be configured to send feature information corresponding to historical behavior information related to the first user in the second service to the server.
  • the recommendation system further includes a third device, and the first device is specifically configured to send the user information of the first user to the server through the third device.
  • the server is specifically configured to send recommendation information to the first device through the third device.
  • the user information of the first user includes: identification information of the first user, and/or identification information of the first user group.
  • the server is further configured to determine the second service according to the characteristic information corresponding to the historical behavior information related to the first user in the second service and the characteristic information corresponding to the historical behavior information of one or more other users in the second service.
  • a user group and, sending identification information of the first user group to the first device.
  • the feature information corresponding to the historical behavior information related to the first user in the second service includes: the hash value corresponding to the historical behavior information related to the first user in the second service.
  • the feature information corresponding to the historical behavior information of the other one or more users in the second service includes: a hash value corresponding to the historical behavior information of the other one or more users in the second service.
  • the hash value corresponding to the historical behavior information related to the first user in the second service is obtained by hashing the first encoding result using the first hash algorithm, and the first encoding result is obtained by using the first encoding
  • the method is obtained by encoding the historical behavior information related to the first user in the second service.
  • the hash value corresponding to the historical behavior information of one or more users in the second service is obtained by hashing the second encoding result using the first hash algorithm, and the second encoding result is obtained by using the first encoding method to It is obtained by encoding the historical behavior information of one or more other users in the second service.
  • the first device is specifically configured to, when detecting that the first user uses the first service for the first time, send the user information of the first user to the server.
  • the recommendation system can realize the functions corresponding to all the steps of the recommendation method described in the first aspect and any possible implementation manner of the first aspect, which will not be repeated here.
  • the present application provides a recommendation method that can be applied to the first device.
  • the method includes: sending user information of the first user to the server; receiving recommendation information from the server; the recommendation information is information related to the use of the first service by users in the first user group; the user information of the first user is related to the second service.
  • a user group is related; the first user group is based on the feature information corresponding to the historical behavior information related to the first user in the second service and the historical behavior information of one or more other users in the second service.
  • the feature information matches what is determined.
  • the method can narrow down the results of user grouping for the first user, effectively improve the accuracy of personalized recommendation, and recommend more interesting recommendation results for the first user.
  • the first device includes a first service and a second service; the method further includes: sending feature information corresponding to historical behavior information related to the first user in the second service to the server.
  • the historical behavior information related to the first user in the second service may include: the corresponding historical behavior information of the first user in the second service, or other accounts (family public account) associated with the first user's account or other user's account) corresponding historical behavior information in the second service.
  • the first device includes the first service
  • the second device includes the second service
  • the first device and the second device may be two terminal devices in a multi-device coordination scenario.
  • the sending the user information of the first user to the server includes: sending the user information of the first user to the server through a third device.
  • the receiving the recommendation information from the server includes: receiving the recommendation information from the server through the third device.
  • the first device may be a device with less device resources and cannot be independently networked and/or human-computer interaction
  • the third device may be a device with relatively rich device resources and independent network and human-computer interaction.
  • the sending to the server the characteristic information corresponding to the historical behavior information related to the first user in the second service may include: sending the historical behavior information related to the first user in the second service to the server through a third device Corresponding feature information.
  • the method further includes: receiving identification information of the first user group sent from the server; the first user group is characteristic information corresponding to the historical behavior information related to the first user in the second service by the server, and other feature information corresponding to the historical behavior information of one or more users in the second service.
  • the sending the user information of the first user to the server includes: sending the user information of the first user to the server when it is detected that the first user uses the first service for the first time.
  • making personalized recommendations to the first user when the first user uses the first service for the first time can be referred to as performing a user cold start for the first user.
  • This method can improve the recommendation for the first user when performing a user cold start. Accuracy, the effect of user cold start can be better.
  • the present application provides a recommendation device, which can be applied to an electronic device (such as the first device), so that the electronic device can realize the above mentioned in the third aspect and any possible implementation manner of the third aspect. recommended method.
  • the functions of the device can be realized by hardware, and can also be realized by executing corresponding software by hardware.
  • the hardware or software includes one or more modules or units corresponding to the steps in the recommended method described in the third aspect and any possible implementation manner of the third aspect.
  • the device may include: a sending unit, a receiving unit, and the like.
  • the sending unit may be used to send user information of the first user to the server;
  • the receiving unit may be used to receive recommendation information from the server;
  • the recommendation information is information related to the use of the first service by users in the first user group;
  • the first user The user information is related to the first user group;
  • the first user group is based on the characteristic information corresponding to the historical behavior information related to the first user in the second service, and the other one or more users in the second service It is determined by matching the characteristic information corresponding to the historical behavior information.
  • the first device includes a first service and a second service; the sending unit is further configured to send feature information corresponding to historical behavior information related to the first user in the second service to the server.
  • the first device includes the first service
  • the second device includes the second service
  • the sending unit is specifically configured to send the user information of the first user to the server through the third device.
  • the receiving unit is specifically configured to receive recommendation information from the server through the third device.
  • the sending unit is specifically configured to send the feature information corresponding to the historical behavior information related to the first user in the second service to the server through the third device.
  • the receiving unit is further configured to receive identification information of the first user group sent from the server; the first user group is characteristic information corresponding to the historical behavior information related to the first user in the second service by the server, and other feature information corresponding to the historical behavior information of one or more users in the second service.
  • the sending unit is specifically configured to send the user information of the first user to the server when it is detected that the first user uses the first service for the first time.
  • the device can realize the functions corresponding to all the steps of the recommendation method described in the third aspect and any possible implementation manner of the third aspect, which will not be repeated here.
  • the present application provides an electronic device, for example, the electronic device may be the above-mentioned first device.
  • the electronic device includes: a processor, a memory for storing processor-executable instructions; when the processor is configured to execute the instructions, the electronic device implements the third aspect and any possible implementation manner of the third aspect. recommended method described above.
  • the present application provides a computer-readable storage medium, on which computer program instructions are stored; when the computer program instructions are executed by an electronic device, the electronic device realizes any one of the third aspect and the third aspect. Recommended method as described in one possible implementation.
  • the present application provides a computer program product, including computer-readable codes, or a non-volatile computer-readable storage medium carrying computer-readable codes, when the computer-readable codes are run in an electronic device
  • the processor in the electronic device implements the recommendation method described in the third aspect and any possible implementation manner of the third aspect.
  • the present application provides a recommendation method that can be applied to a server.
  • the method includes: receiving user information of the first user sent by the first device; and sending recommendation information to the first device.
  • the recommended information is information related to the use of the first service by users in the first user group; the user information of the first user is related to the first user group; the first user group is based on the information related to the first user in the second service.
  • the characteristic information corresponding to the historical behavior information of the user and the characteristic information corresponding to the historical behavior information of the other one or more users in the second service are determined by matching.
  • the method determines that the first user is in User grouping results in the second service, and based on the user grouping results of the first user in the second service, personalized recommendations are made to the first user in the first service, which can narrow down the results of user grouping for the first user, Effectively improve the accuracy of personalized recommendation, and recommend more interesting recommendation results for the first user.
  • the first device includes a first service and a second service; the method further includes: receiving characteristic information corresponding to historical behavior information related to the first user in the second service sent by the first device .
  • both the first service and the second service may be provided by the first device.
  • the historical behavior information related to the first user in the second service may include: the corresponding historical behavior information of the first user in the second service, or other accounts (family public account) associated with the first user's account or other user's account) corresponding historical behavior information in the second service.
  • the first device includes a first service
  • the second device includes a second service
  • the method further includes: receiving historical behavior related to the first user in the second service from the second device The feature information corresponding to the information.
  • the first service may be provided by the first device, and the second service may be provided by the second device.
  • the first device and the second device may be two terminal devices in a multi-device coordination scenario.
  • the receiving the user information of the first user sent by the first device includes: receiving the user information of the first user sent by the first device through a third device.
  • the sending the recommendation information to the first device includes: sending the recommendation information to the first device through the third device.
  • the first device may be a device with less device resources and cannot be independently networked and/or human-computer interaction
  • the third device may be a device with relatively rich device resources and independent network and human-computer interaction.
  • the receiving the feature information corresponding to the historical behavior information related to the first user in the second service sent by the first device may include: receiving the feature information from the second service sent by the first device through the third device. Feature information corresponding to historical behavior information related to the first user.
  • the receiving the feature information corresponding to the historical behavior information related to the first user in the second service sent by the second device may include: receiving, by the fourth device, the feature information corresponding to the historical behavior information in the second service sent by the second device. Feature information corresponding to historical behavior information related to the first user.
  • the second device may be a device with less device resources, which cannot be independently networked and/or human-computer interaction
  • the fourth device may be a device with relatively rich device resources, which may be independently networked and human-computer interaction.
  • the method further includes: according to the characteristic information corresponding to the historical behavior information related to the first user in the second service and the characteristic information corresponding to the historical behavior information of one or more other users in the second service, Determine the first user group; send the identification information of the first user group to the first device.
  • the first user uses the first service for the first time.
  • making personalized recommendations to the first user when the first user uses the first service for the first time can be referred to as performing a user cold start for the first user.
  • This method can improve the recommendation for the first user when performing a user cold start. Accuracy, the effect of user cold start can be better.
  • the present application provides a recommendation device, which can be applied to an electronic device (such as a server), so that the electronic device implements the recommendation method described in the eighth aspect and any possible implementation of the eighth aspect .
  • the functions of the device can be realized by hardware, and can also be realized by executing corresponding software by hardware.
  • the hardware or software includes one or more modules or units corresponding to the steps in the recommended method described in the eighth aspect and any possible implementation manner of the eighth aspect.
  • the device may include: a receiving unit and a sending unit.
  • the receiving unit may be configured to receive user information of the first user sent from the first device;
  • the sending unit may be configured to send recommendation information to the first device.
  • the recommended information is information related to the use of the first service by users in the first user group;
  • the user information of the first user is related to the first user group;
  • the first user group is based on the information related to the first user in the second service
  • the characteristic information corresponding to the historical behavior information of the user and the characteristic information corresponding to the historical behavior information of the other one or more users in the second service are determined by matching.
  • the first device includes a first service and a second service; the receiving unit is further configured to receive characteristic information corresponding to historical behavior information related to the first user in the second service sent by the first device .
  • the first device includes a first service
  • the second device includes a second service
  • the receiving unit is further configured to receive historical behavior related to the first user in the second service sent from the second device The feature information corresponding to the information.
  • the receiving unit is specifically configured to receive, through the third device, the user information of the first user sent from the first device.
  • the sending unit is specifically configured to send recommendation information to the first device through the third device.
  • the receiving unit is specifically configured to receive, through the third device, feature information corresponding to historical behavior information related to the first user in the second service sent by the first device.
  • the receiving unit is specifically configured to receive, through the fourth device, feature information corresponding to historical behavior information related to the first user in the second service sent by the second device.
  • the device further includes: a processing unit.
  • the processing unit may be configured to determine the first user group according to the feature information corresponding to the historical behavior information related to the first user in the second service and the feature information corresponding to the historical behavior information of one or more other users in the second service Group.
  • the sending unit is further configured to send the identification information of the first user group to the first device.
  • the device can cooperate to realize the functions corresponding to all the steps of the recommendation method described in the eighth aspect and any possible implementation manner of the eighth aspect, which will not be repeated here.
  • the present application provides an electronic device, for example, the electronic device may be the above-mentioned server.
  • the electronic device includes: a processor, a memory for storing processor-executable instructions; when the processor is configured to execute the instructions, the electronic device implements the eighth aspect and any possible implementation manner of the eighth aspect. recommended method described above.
  • the present application provides a computer-readable storage medium, on which computer program instructions are stored; when the computer program instructions are executed by an electronic device, the electronic device realizes any of the eighth and eighth aspects. Recommended method as described in one possible implementation.
  • the present application provides a computer program product, including computer-readable codes, or a non-volatile computer-readable storage medium bearing computer-readable codes, when the computer-readable codes run in an electronic device
  • the processor in the electronic device implements the recommended method described in the eighth aspect and any possible implementation manner of the eighth aspect.
  • the servers described in the first to twelfth aspects above may be server clusters or cloud servers, or the servers may be divided into two servers, such as a recommendation server and a user portrait server.
  • FIG. 1 is a schematic diagram of the composition of the recommendation system provided by the embodiment of the present application.
  • FIG. 2 is a schematic structural diagram of a terminal device 110 provided in an embodiment of the present application.
  • Fig. 3 is a schematic flow chart of the recommendation method provided by the embodiment of the present application.
  • Fig. 4 is another schematic flowchart of the recommendation method provided by the embodiment of the present application.
  • Fig. 5 is another schematic flowchart of the recommendation method provided by the embodiment of the present application.
  • FIG. 6 is a schematic diagram of an application scenario provided by an embodiment of the present application.
  • FIG. 7 is a schematic structural diagram of a recommendation device provided in an embodiment of the present application.
  • FIG. 8 is another schematic structural diagram of a recommendation device provided by an embodiment of the present application.
  • references to "one embodiment” or “some embodiments” or the like in this specification means that a particular feature, structure, or characteristic described in connection with the embodiment is included in one or more embodiments of the present application.
  • appearances of the phrases “in one embodiment,” “in some embodiments,” “in other embodiments,” “in other embodiments,” etc. in various places in this specification are not necessarily All refer to the same embodiment, but mean “one or more but not all embodiments” unless specifically stated otherwise.
  • the terms “including”, “comprising”, “having” and variations thereof mean “including but not limited to”, unless specifically stated otherwise.
  • the term “connected” includes both direct and indirect connections, unless otherwise stated.
  • first and second are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of indicated technical features. Thus, a feature defined as “first” and “second” may explicitly or implicitly include one or more of these features.
  • Personalized recommendation generally refers to discovering the user's individual needs and interest characteristics by analyzing and mining the user's historical behavior information, and recommending information or products that the user may be interested in to the user.
  • Personalized recommendations can be applied to e-commerce systems, social networks, advertising recommendations, search engines, and many other aspects.
  • the e-commerce platform can include a recommendation system dedicated to personalized recommendation. are more interested, and recommend products that users are more interested in.
  • the video platform can also include a recommendation system dedicated to personalized recommendation. Types of videos are more interesting and recommend videos that are more interesting to users.
  • the recommendation system can also evaluate the similarity between the user and other users, and based on the user collaborative filtering (user collaboration filter, UserCF) algorithm, recommend to the user videos that are similar to the user and other users are interested in.
  • the movies that user A has watched many times are all sci-fi movies, such as: sci-fi movie 1, sci-fi movie 2, etc.
  • the recommendation system finds that user B has also watched sci-fi movie 1 and sci-fi movie 2 many times through data analysis, and User B has also watched science fiction movie 3 many times, then the recommendation system can recommend science fiction movie 3 to user A.
  • personalized recommendation often faces incremental problems, such as: personalized recommendation for new users.
  • new users do not have historical behavior information for the recommendation system of personalized recommendation, and the recommendation system cannot rely on the user's historical behavior information to make personalized recommendations to users.
  • an e-commerce platform when a new user registers and logs in to the e-commerce platform, this new user is a new user.
  • the e-commerce platform does not record the historical shopping behavior of the new user.
  • the recommendation system of the e-commerce platform cannot recommend products to the user based on the user's historical shopping behavior.
  • the recommendation system can adopt some "user cold start” strategies to make personalized recommendations to users, and in the process of users using services (such as: e-commerce services provided by e-commerce platforms, video viewing services provided by video platforms, etc.)
  • services such as: e-commerce services provided by e-commerce platforms, video viewing services provided by video platforms, etc.
  • the user s historical behavior information is accumulated and collected. After a certain amount of user’s historical behavior information is collected, the recommendation can be completed in accordance with the aforementioned method of analyzing and mining the user’s historical behavior information to make personalized recommendations to the user.
  • the process of accumulating the historical behavior information of the user from 0 to 1 is the process of cold start.
  • the common "user cold start” strategy is: the recommendation system collects some known user information of the user, such as: the device information of the device used by the user, the location of the device, etc.; using these known user information, the recommendation system can After grouping, the recommendation system can use the UserCF algorithm (or collaborative recommendation algorithm) to make personalized recommendations for users in the same group.
  • the UserCF algorithm or collaborative recommendation algorithm
  • the recommendation system of the e-commerce platform can collect the device information and device location of the mobile phone used by the new user , such as assuming that the device information of the mobile phone used by the new user includes: the brand of the mobile phone is brand P, and the location of the device is city X. Then, the recommendation system can group the new user with other users on the e-commerce platform according to the device information and device location of the mobile phone used by the new user. For example, the recommendation system can attribute the new user to To the "Brand P+X City" crowd. Afterwards, the recommendation system can recommend to the new user products that are of interest to other users in the "Brand P+X City” crowd based on the UserCF algorithm.
  • the current "user cold start” strategy can obtain limited known user information, the user groups are too large, and the effect of personalized recommendation is mediocre. For example, users may not be interested in the recommendation results of personalized recommendations.
  • the embodiment of the present application provides a recommendation method, which can be applied to the scenario of performing personalized recommendation in the first service (the service can also be called business, field, etc.).
  • the first service may include the function of making personalized recommendations to users
  • the recommendation method provided in the embodiment of this application may be applied to the function of making personalized recommendations to users in the first service.
  • the method can use the historical behavior information of the new user in the second service to perform knowledge transfer, and determine that the new user in the second service Based on the user grouping results of the new user in the second service, and based on the user grouping results of the new user in the second service, collaborative recommendation is made to the new user in the first service, so as to realize the user cold start of the new user in the first service.
  • the first service and the second service are different.
  • This method can narrow down the results of user grouping, effectively improve the accuracy of personalized recommendation, and recommend more interesting recommendation results for new users, and the effect of user cold start can be better.
  • the function of making personalized recommendations to users included in the first service may be called “guess what you like”, “relevant recommendations”, “personalized recommendations”, “user subscribes to personalized recommendations”, “search Intent recognition” and so on, there is no limitation on the name of the function of personalized recommendation.
  • the first service may refer to a single service, for example, a search service (for example, a search platform can provide users with a search information service through a browser or other applications), an information flow service (for example, a browser can use Information flow services provide users with browseable news, information and other information flows), e-commerce services (for example, e-commerce platforms can provide users with services to purchase goods through some applications (Apps), music platforms provide services for users through some applications. Music services provided by users (such as listening to songs), video services provided by video platforms to users through some applications (such as watching movies or short videos), reading services provided by reading platforms to users through some applications (such as e-book reading), etc.
  • a search service for example, a search platform can provide users with a search information service through a browser or other applications
  • an information flow service for example, a browser can use
  • Information flow services provide users with browseable news, information and other information flows
  • e-commerce services for example, e-commerce platforms can provide users with services to purchase goods through some applications (Apps
  • the second service may also include the aforementioned search service, information flow service, e-commerce service, music service, video service, reading service, etc., but the second service is different from the first service.
  • the first service is information flow service
  • the second service is search service.
  • a browser is installed on the mobile phone of user 1, and the services provided in the browser include: search service and information streaming service.
  • User 1 has used search services before, and there is a lot of historical search behavior information (such as search terms, clicks on search result pages, etc.), but user 1 has never used the information flow service (that is, has not read the information flow).
  • For the information flow service when user 1 uses the information flow service for the first time, user 1 is a new user. However, since user 1 is using the information flow service for the first time, the personalized recommendation system related to the information flow service does not know the user's interests and preferences, and cannot make recommendations based on user 1's historical behavior information in the information flow service.
  • the recommendation method provided by the embodiment of the present application can use the historical behavior information of user 1 in the search service to carry out knowledge transfer, determine the user grouping results of user 1 in the search service, and then, based on the user grouping of user 1 in the search service As a result, information streams that other users in the same group are interested in are recommended to user 1, thereby solving the cold start problem of user 1 when using the information stream service for the first time.
  • the second service may include one or more types, which is not limited in this application.
  • the first service is an information flow service
  • the second service may include a search service, a reading service, a video service, and the like.
  • the above examples of the first service and the second service are only partial examples, and the present application does not limit the specific types of the first service and the second service.
  • the first service may generally refer to a service provided by a certain application, and does not refer to a single service (such as the above-mentioned information flow service, search service, etc.), and the second service may refer to a service provided by other applications.
  • the first service may be a service provided by a browser, which may specifically include various services such as search service, information streaming service, and video service;
  • the second service may be a service provided by a certain reading app, which may specifically include e-book reading service .
  • the application that provides the first service is different from the application that provides the second service, and it can also be considered that the first service is different from the second service.
  • browser A can provide user 1 with various services such as search service, information flow service, reading service, and video service
  • short video application B User 1 can be provided with a short video browsing service.
  • User 1 has used the service provided by browser A (that is, the second service) before, and there is a lot of historical behavior information in browser A (such as using search services, reading information streams, etc.), but user 1 has not used short
  • the service provided by the video application B ie the first service.
  • For the service provided by short video application B when user 1 uses the service provided by short video application B for the first time, user 1 is a new user.
  • the personalized recommendation system related to the service provided by short video application B does not know the user's interests and preferences, and cannot use the service provided by user 1 in short video application B. Recommendations based on historical behavior information in the service.
  • the recommendation method provided by the embodiment of the present application can use the historical behavior information of user 1 in the service provided by browser A to carry out knowledge transfer, determine the user grouping result of user 1 in the service provided by browser A, and then, based on the user 1 In the results of user grouping in the service provided by browser A, recommend short videos that other users in the same group are interested in to user 1, so as to solve the cold start when user 1 uses the service provided by short video application B for the first time question.
  • the second service may include one or more services provided by applications, that is, the second service may include one or more services, which is not limited in this application.
  • the first service is a service provided by application A
  • the second service may include a service provided by application B, a service provided by application C, and so on.
  • the embodiment of the present application does not limit the relationship between the first service and the second service.
  • the first service and the second service may be two different types of services (such as search service and information flow service) as described in the foregoing embodiments, or services provided by different applications.
  • the first service and the second service may also be two services of the same type but with different service channels.
  • both the first service and the second service are search services, but the first service uses search engine 1 to provide search services, and the second service uses search engine 2 to provide services.
  • both the first service and the second service are video services, but the first service is a video service provided by video platform 1, and the second service is a video service provided by video platform 2, etc.
  • the first service and the second service can also be two different services provided by the same application (such as the search service and the information flow service provided by the browser), or the first service and the second service It can be two different services provided by the same service provider (for example, service provider A provides reading service and information flow service).
  • user 1 may have the same user account, such as account 1, in the first service and the second service.
  • the historical behavior information of the user 1 in the second service may refer to: the historical behavior information corresponding to the account 1 in the second service, that is, the user 1 has used the second service through the account 1.
  • the collaborative recommendation of user grouping results to user 1 in the first service may include: performing knowledge transfer using historical behavior information corresponding to account 1 in the second service, determining the user grouping results of account 1 in the second service, and based on The user grouping results of the account 1 in the second service are collaboratively recommended to the account 1 in the first service.
  • user 1 may have different user accounts in the first service and the second service, user 1 has account 1 in the first service, and user 1 in the second service may have different user accounts.
  • the second service has account 2.
  • the historical behavior information of the user 1 in the second service may refer to: the historical behavior information corresponding to the account 2 in the second service, that is, the user 1 has used the second service through the account 2.
  • the collaborative recommendation of user grouping results to user 1 in the first service may include: performing knowledge transfer using historical behavior information corresponding to account 2 in the second service, determining the user grouping results of account 2 in the second service, and based on The user grouping results of account 2 in the second service are collaboratively recommended to account 1 in the first service.
  • user 1 has account 1 in the first service, but does not have an account in the second service, but there is an account associated with account 1 in the second service.
  • account number 2 may be the account of the user 2, but the account 2 and the account 1 have association relations such as binding and sharing.
  • account 2 may be a public family account of user 1, but account 2 is associated with account 1, or account 1 may be a public family account of user 1, account 2 may be an account of user 2, but account 2 is related to account 1.
  • Account 1 has an association relationship, etc.
  • the recommendation method provided in the embodiment of this application can also use the account 2 Perform knowledge transfer on the corresponding historical behavior information in the second service, determine the user grouping results of account 2 in the second service, and transfer to account 1 in the first service based on the user grouping results of account 2 in the second service Collaborative recommendation.
  • the historical behavior information corresponding to account 2 in the second service can also be considered as the historical behavior information related to user 1 in the second service
  • the user grouping result of account 2 in the second service can be considered as user 1 in the second service.
  • the historical behavior information corresponding to the user 1 in the second service described in the foregoing embodiments can also be considered as the historical behavior information related to the user 1 in the second service.
  • the recommendation method provided by the embodiment of the present application can use the historical behavior information related to the user in the second service to carry out knowledge transfer, determine the user grouping results of the user in the second service, and based on the user's information in the second service The user grouping result is collaboratively recommended to the user in the first service.
  • the historical behavior information related to the user in the second service may include: the corresponding historical behavior information of the user in the second service, or other accounts associated with the user's account (family public account or other user's account) in the second service.
  • the corresponding historical behavior information in the service includes: the historical behavior information corresponding to the user in the second service is taken as an example for illustration, which is not intended to be limiting.
  • FIG. 1 is a schematic diagram of the composition of the recommendation system provided in the embodiment of the present application.
  • the recommendation system provided by the embodiment of the present application may include: a terminal device 110 , a recommendation server 120 , and a user portrait server 130 .
  • the terminal device 110, the recommendation server 120, and the user portrait server 130 may be connected to each other through a wired network or a wireless network.
  • the terminal device 110 may be a mobile phone, a tablet computer, a wearable device, a vehicle-mounted device, an augmented reality (augmented reality, AR)/virtual reality (virtual reality, VR) device, a notebook computer, an ultra mobile personal computer (ultra -mobile personal computer, UMPC), netbook, personal digital assistant (personal digital assistant, PDA), etc., the embodiment of the present application does not limit the specific type of the terminal device 110.
  • FIG. 2 is a schematic structural diagram of the terminal device 110 provided in the embodiment of the present application.
  • the terminal device 110 may include: a processor 210, an external memory interface 220, an internal memory 221, a universal serial bus (universal serial bus, USB) interface 230, a charging management module 240, a power management module 241, a battery 242, antenna 1, antenna 2, mobile communication module 250, wireless communication module 260, audio module 270, speaker 270A, receiver 270B, microphone 270C, earphone jack 270D, sensor module 280, button 290, motor 291, indicator 292, camera 293, a display screen 294, and a subscriber identification module (subscriber identification module, SIM) card interface 295, etc.
  • SIM subscriber identification module
  • the processor 210 may include one or more processing units, for example: the processor 210 may include an application processor (application processor, AP), a modem processor, a graphics processing unit (graphics processing unit, GPU), an image signal processor (image signal processor, ISP), controller, memory, video codec, digital signal processor (digital signal processor, DSP), baseband processor, and/or neural network processor (neural-network processing unit, NPU) wait. Wherein, different processing units may be independent devices, or may be integrated in one or more processors.
  • application processor application processor, AP
  • modem processor graphics processing unit
  • GPU graphics processing unit
  • image signal processor image signal processor
  • ISP image signal processor
  • controller memory
  • video codec digital signal processor
  • DSP digital signal processor
  • baseband processor baseband processor
  • neural network processor neural-network processing unit, NPU
  • the controller may be the nerve center and command center of the electronic equipment.
  • the controller can generate an operation control signal according to the instruction opcode and timing signal, and complete the control of fetching and executing the instruction.
  • a memory may also be provided in the processor 210 for storing instructions and data.
  • the memory in processor 210 is a cache memory.
  • the memory may hold instructions or data that the processor 210 has just used or recycled.
  • processor 210 may include one or more interfaces.
  • the interface may include an integrated circuit (inter-integrated circuit, I2C) interface, an integrated circuit built-in audio (inter-integrated circuit sound, I2S) interface, a pulse code modulation (pulse code modulation, PCM) interface, a universal asynchronous transmitter (universal asynchronous receiver/transmitter, UART) interface, mobile industry processor interface (mobile industry processor interface, MIPI), general-purpose input and output (general-purpose input/output, GPIO) interface, subscriber identity module (subscriber identity module, SIM) interface, and /or universal serial bus (universal serial bus, USB) interface, etc.
  • I2C integrated circuit
  • I2S integrated circuit built-in audio
  • PCM pulse code modulation
  • PCM pulse code modulation
  • UART universal asynchronous transmitter
  • MIPI mobile industry processor interface
  • GPIO general-purpose input and output
  • subscriber identity module subscriber identity module
  • SIM subscriber identity module
  • USB universal serial bus
  • the interface connection relationship between the modules shown in this embodiment is only a schematic illustration, and does not constitute a structural limitation of the terminal device 110 .
  • the terminal device 110 may also adopt different interface connection modes in the foregoing embodiments, or a combination of multiple interface connection modes.
  • the charging management module 240 is configured to receive charging input from the charger. While the charging management module 240 is charging the battery 242 , it can also supply power to the electronic device through the power management module 241 .
  • the power management module 241 is used for connecting the battery 242 , the charging management module 240 and the processor 210 .
  • the power management module 241 receives the input from the battery 242 and/or the charging management module 240 to provide power for the processor 210 , internal memory 221 , external memory, display screen 294 , camera 293 , and wireless communication module 260 .
  • the power management module 241 can also be used to monitor parameters such as battery capacity, battery cycle times, and battery health status (leakage, impedance).
  • the power management module 241 can also be set in the processor 210 .
  • the power management module 241 and the charging management module 240 may also be set in the same device.
  • the wireless communication function of the electronic device can be realized by the antenna 1, the antenna 2, the mobile communication module 250, the wireless communication module 260, the modem processor and the baseband processor.
  • the mobile communication module 250 can provide wireless communication solutions including 2G/3G/4G/5G applied on the terminal device 110 .
  • the mobile communication module 250 may include at least one filter, switch, power amplifier, low noise amplifier (low noise amplifier, LNA) and the like.
  • the mobile communication module 250 can receive electromagnetic waves through the antenna 1, filter and amplify the received electromagnetic waves, and send them to the modem processor for demodulation.
  • the mobile communication module 250 can also amplify the signal modulated by the modem processor, convert it into electromagnetic wave and radiate it through the antenna 1 .
  • at least part of the functional modules of the mobile communication module 250 may be set in the processor 210 .
  • at least part of the functional modules of the mobile communication module 250 and at least part of the modules of the processor 210 may be set in the same device.
  • a modem processor may include a modulator and a demodulator.
  • the modem processor may be a stand-alone device.
  • the modem processor may be independent of the processor 210, and be set in the same device as the mobile communication module 250 or other functional modules.
  • the wireless communication module 260 can provide wireless local area networks (wireless local area networks, WLAN) (such as wireless fidelity (Wireless Fidelity, Wi-Fi) network), bluetooth (bluetooth, BT), global navigation satellite, etc. applied on the terminal device 110.
  • System global navigation satellite system, GNSS
  • frequency modulation frequency modulation, FM
  • near field communication technology near field communication, NFC
  • infrared technology infrared, IR
  • the wireless communication module 260 may be one or more devices integrating at least one communication processing module.
  • the wireless communication module 260 receives electromagnetic waves via the antenna 2 , frequency-modulates and filters the electromagnetic wave signals, and sends the processed signals to the processor 210 .
  • the wireless communication module 260 can also receive the signal to be sent from the processor 210 , frequency-modulate it, amplify it, and convert it into electromagnetic waves through the antenna 2 to radiate out.
  • the antenna 1 of the terminal device 110 is coupled to the mobile communication module 250, and the antenna 2 is coupled to the wireless communication module 260, so that the terminal device 110 can communicate with the network and other devices (such as the recommendation server 120, user portrait Server 130) communicates.
  • the wireless communication technology may include global system for mobile communications (GSM), general packet radio service (general packet radio service, GPRS), code division multiple access (code division multiple access, CDMA), broadband Code division multiple access (wideband code division multiple access, WCDMA), time division code division multiple access (time-division code division multiple access, TD-SCDMA), long term evolution (long term evolution, LTE), BT, GNSS, WLAN, NFC , FM, and/or IR techniques, etc.
  • GSM global system for mobile communications
  • general packet radio service general packet radio service
  • CDMA code division multiple access
  • WCDMA broadband Code division multiple access
  • time division code division multiple access time-division code division multiple access
  • TD-SCDMA time-division code division multiple access
  • the GNSS may include a global positioning system (global positioning system, GPS), a global navigation satellite system (global navigation satellite system, GLONASS), a Beidou navigation satellite system (beidou navigation satellite system, BDS), a quasi-zenith satellite system (quasi -zenith satellite system (QZSS) and/or satellite based augmentation systems (SBAS).
  • the terminal device 110 can use the wireless communication module 260 to establish a wireless connection with the recommendation server 120 and the user portrait server 130 through wireless communication technology. Based on the established wireless connection, the terminal device 110 can send information or messages to the recommendation server 120 and the user portrait server 130 , and can also receive information or messages from the recommendation server 120 and the user portrait server 130 .
  • the terminal device 110 implements a display function through a GPU, a display screen 294, an application processor, and the like.
  • the GPU is a microprocessor for image processing, and is connected to the display screen 294 and the application processor.
  • Processor 210 may include one or more GPUs that execute program instructions to generate or change display information.
  • the display screen 294 is used to display images, videos and the like.
  • Display 294 includes a display panel.
  • the display panel can be a liquid crystal display (LCD), an organic light-emitting diode (OLED), an active matrix organic light emitting diode or an active matrix organic light emitting diode (active-matrix organic light emitting diode, AMOLED), flexible light-emitting diode (flex light-emitting diode, FLED), Miniled, MicroLed, Micro-oLed, quantum dot light emitting diodes (quantum dot light emitting diodes, QLED), etc.
  • the terminal device 110 may include 1 or N display screens 294, where N is a positive integer greater than 1.
  • the terminal device 110 may implement a shooting function through an ISP, a camera 293 , a video codec, a GPU, a display screen 294 , an application processor, and the like.
  • the ISP is used for processing the data fed back by the camera 293 .
  • the ISP may be located in the camera 293 .
  • Camera 293 is used to capture still images or video.
  • the terminal device 110 may include 1 or N cameras 293, where N is a positive integer greater than 1.
  • Video codecs are used to compress or decompress digital video.
  • An electronic device may support one or more video codecs.
  • the external memory interface 220 may be used to connect an external memory card, such as a Micro SD card, to expand the storage capacity of the terminal device 110.
  • the external memory card communicates with the processor 210 through the external memory interface 220 to implement a data storage function. Such as saving music, video and other files in the external memory card.
  • the internal memory 221 may be used to store computer-executable program codes including instructions.
  • the processor 210 executes various functional applications and data processing of the terminal device 110 by executing instructions stored in the internal memory 221 .
  • the processor 210 can establish a connection with the recommendation server 120 and the user portrait server 130 through the wireless communication module 260 by executing instructions stored in the internal memory 221, and communicate with the recommendation server 120 and the user portrait server. 130 performs data interaction.
  • the internal memory 221 may include an area for storing programs and an area for storing data. Wherein, the stored program area can store an operating system, at least one application program required by a function (such as a sound playing function, an image playing function, etc.) and the like.
  • the storage data area can store data (such as audio data, phone book, etc.) created during the use of the electronic device.
  • the internal memory 221 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one magnetic disk storage device, flash memory device, universal flash storage (universal flash storage, UFS) and the like.
  • the terminal device 110 may implement an audio function through an audio module 270, a speaker 270A, a receiver 270B, a microphone 270C, an earphone interface 270D, and an application processor. Such as calls, music playback, recording, etc.
  • the audio module 270 is used to convert digital audio information into analog audio signal output, and is also used to convert analog audio input into digital audio signal.
  • the audio module 270 may also be used to encode and decode audio signals.
  • the audio module 270 can be set in the processor 210 , or some functional modules of the audio module 270 can be set in the processor 210 .
  • Speaker 270A also referred to as a "horn" is used to convert audio electrical signals into sound signals.
  • Terminal device 110 can listen to music through speaker 270A, or listen to hands-free calls.
  • Receiver 270B also called “earpiece” is used to convert audio electrical signals into audio signals.
  • the receiver 270B can be placed close to the human ear to receive the voice.
  • the microphone 270C also called “microphone” or “microphone” is used to convert sound signals into electrical signals.
  • the user can make a sound by approaching the microphone 270C with a human mouth, and input the sound signal to the microphone 270C.
  • the electronic device may be provided with at least one microphone 270C.
  • the electronic device can be provided with two microphones 270C, which can also implement a noise reduction function in addition to collecting sound signals.
  • the terminal device 110 can also be provided with three, four or more microphones 270C to realize sound signal collection, noise reduction, identify sound sources, realize directional recording functions, and the like.
  • the earphone interface 270D is used for connecting wired earphones.
  • the earphone interface 270D can be a USB interface 230, or a 3.5mm open mobile terminal platform (OMTP) standard interface, or a cellular telecommunications industry association of the USA (CTIA) standard interface.
  • OMTP open mobile terminal platform
  • CTIA cellular telecommunications industry association of the USA
  • the sensor module 280 may include a pressure sensor 280A, a gyro sensor 280B, an air pressure sensor 280C, a magnetic sensor 280D, an acceleration sensor 280E, a distance sensor 280F, a proximity light sensor 280G, a fingerprint sensor 280H, a temperature sensor 280J, a touch sensor 280K, an ambient light sensor 280L, bone conduction sensor 280M, etc.
  • the structure of the terminal device 110 shown in FIG. 2 does not constitute a specific limitation on the terminal device 110 .
  • the terminal device 110 when the terminal device 110 is a tablet computer, desktop, laptop, handheld computer, notebook computer, UMPC, netbook, and other devices such as cellular phones, PDAs, and AR/VR devices, the terminal device 110 More or fewer components than shown in FIG. 2 may be included, or some components may be combined, or some components may be split, or a different arrangement of components may be included.
  • the components shown in Fig. 2 can be implemented in hardware, software or a combination of software and hardware.
  • the embodiment of the present application does not limit the specific structure of the terminal device 110 .
  • the recommendation server 120 and the user portrait server 130 are represented separately, but in some possible embodiments, the recommendation server 120 and the user portrait server 130 may also be the same The server, alternatively, the recommendation server 120 and the user portrait server 130 may also be the same server cluster or cloud server, which is not limited here.
  • the terminal device 110 can provide users with services through Apps, for example, a first App can be installed on the terminal device 110, and the first App can provide users with first services, second services, etc.
  • a first App and a second App may be installed on the terminal device 110, the first App may provide the user with the first service, and the second App may provide the user with the second service, and so on.
  • the terminal device 110 may implement a personalized recommendation function in the service based on the recommendation system shown in FIG. 1 above.
  • the terminal device 110 may record or collect the user's Historical behavior information in the second service.
  • the user portrait server 130 can calculate and analyze the user's interest in the second service according to the user's historical behavior information in the second service, and understand the user's interest and preference in the second service.
  • the recommendation server 120 may recommend service content of the second service that the user is interested in according to the result of calculation and analysis of the user's interest in the second service by the user portrait server 130 .
  • the second service may be an information flow service
  • the user portrait server 130 calculates and analyzes the user's interest in the second service to obtain that the user is more interested in games
  • the recommendation server 120 may recommend game-related information flow to the user, such as Information.
  • the specific algorithm for recommendation is not limited.
  • the user portrait server 130 can also perform clustering and grouping (that is, user grouping) on the user and other users according to the historical behavior information of the user in the second service and the historical behavior information of other users in the second service, and obtain one or Multiple user groups.
  • the recommendation server 120 may also recommend to the user the service content of the second service that other users in the same user group as the user are interested in according to the user grouping result of the user portrait server 130 . That is, the recommendation server 120 may also perform collaborative recommendation to the user in the second service according to the user grouping result of the user portrait server 130 in the second service.
  • the service content of the second service that other users are interested in may be the service content of the second service that other users have browsed, or it is calculated and analyzed according to the user portrait server 130 on other users' interests in the second service As a result, the service content of the second service recommended to other users.
  • the recommendation server 120 may recommend to the user other users who are in the same user group as the user according to the result of user grouping by the user portrait server 130 in the second service.
  • the service content of the first service that other users are interested in may be the service content of the first service that other users have browsed, or it is calculated and analyzed according to the user portrait server 130 on other users' interests in the first service As a result, the service content of the first service recommended to other users.
  • the above-mentioned historical behavior information may include the specific service content browsed by the user in the process of using the service, and for one or more services
  • This application does not limit the specific content of historical behavior information such as the number of clicks, viewing times, sharing times, browsing time, negative feedback (such as complaints or feedback not interested) and other related information of content (such as a certain group of information streams).
  • historical behavior information may include information such as addresses of webpages browsed by users (referred to as URLs), keywords searched by users, clicked pages searched by users, and the duration of webpages viewed by users.
  • the terminal device 110 after the terminal device 110 records or collects the user's historical behavior information in the second service, it may not need to upload or send the user's historical behavior information in the second service to the user portrait server 130 for user portrait
  • the server 130 calculates and analyzes the user's interest in the second service according to the historical behavior information of the user in the second service.
  • the terminal device 110 After recording or collecting the user's historical behavior information in the second service, the terminal device 110 can first encode the user's historical behavior information in the second service, and perform hash calculation on the encoded result to obtain the user The hash value corresponding to the historical behavior information in the second service.
  • the terminal device 110 may send the hash value corresponding to the historical behavior information of the user in the second service to the user portrait server 130, and the user portrait server 130 may, according to the hash value corresponding to the historical behavior information of the user in the second service, Carry out user grouping for users, and determine the user group to which the user belongs.
  • Different user groups have different group identifiers (such as group IDs), which may also be called group identifiers.
  • the user portrait server 130 may send to the terminal device 110 the group identification of the user group the user belongs to, and the group identification may be referred to as the group identification of the user group the user belongs to in the second service.
  • the grouping identifier of the user group in which the user belongs in the second service is the result of user grouping performed by the user portrait server 130 in the second service in this application.
  • the terminal device 110 can send to the recommendation server 120 the group identification of the user group in which the user belongs in the second service;
  • the group identifier of the group sends to the terminal device 110 the service content of the first service that other users in the same user group as the user in the second service are interested in.
  • the terminal device 110 may recommend to the user the service content of the first service that is of interest to other users in the same user group as the user in the second service as a personalized recommendation result.
  • the terminal device 110 uploads or sends the hash value corresponding to the user's historical behavior information in the second service to the user portrait server 130, and the user portrait server 130 based on the hash value corresponding to the user's historical behavior information in the second service Hive, grouping users into user groups can better protect user privacy.
  • the user portrait server 130 performs user grouping on the user according to the hash value corresponding to the historical behavior information of the user in the second service, and can narrow down the result of user grouping.
  • the terminal device 110 recommends to the user the service content of the first service that is of interest to other users in the same user group as the user in the second service as a personalized recommendation result, which not only realizes Use the knowledge transfer in other services (second service) to perform user cold start in the first service, and effectively improve the accuracy of personalized recommendation, and can recommend more interesting recommendation results for new users, and user cold start The effect can be better.
  • the service content of the first service that other users in the same user group as the user in the second service are interested in may be referred to as recommendation information, and the recommendation information is related to the user group that the user belongs to in the second service.
  • Figure 3 shows the recommendation method provided by the embodiment of the present application schematic diagram of the process. As shown in Fig. 3, the recommendation method may include S301-S309.
  • the terminal device 110 acquires historical behavior information of a first user in a second service.
  • the first user can use the search service to search on the terminal device 110, and the historical behavior information of the first user in the second service may include: The search term searched, the number of searches for the search term, etc.
  • the terminal device 110 encodes the historical behavior information of the first user in the second service by using the first encoding method, and obtains an encoding result corresponding to the historical behavior information of the first user in the second service.
  • the first encoding manner may include one-hot encoding (encoding).
  • One-hot encoding is also known as one-bit effective encoding. It mainly uses N (N is an integer greater than 1) bit status register to encode N states. Each state has an independent register bit, and at any time only one.
  • the encoding result of one-hot encoding may be an N-dimensional vector, in which only one dimension has a value of 1, and the values of other dimensions are all 0.
  • the encoding results corresponding to different historical behavior information may be different, and for each piece of historical behavior information, the encoding result may be unique.
  • the second service as a search service as an example, it is assumed that the historical behavior information of the first user in the second service includes three search websites, namely: website A, website B, and website C.
  • the terminal device 110 can respectively encode the URL A, the URL B, and the URL C in a one-hot encoding manner, and the encoding results corresponding to the URL A, the URL B, and the URL C respectively can be as follows:
  • the encoding result corresponding to each URL can be an 8-dimensional vector (or as an 8-bit array), but each URL corresponds to In the encoding result, only one dimension has a value of 1, and the values of the other dimensions are all 0.
  • the value of the second dimension in the encoding result corresponding to URL A is 1
  • the value of the third dimension in the encoding result corresponding to URL B is 1
  • the value of the sixth dimension in the encoding result corresponding to URL C is 1.
  • an encoding result as an 8-dimensional vector or an 8-dimensional array as an example, and there is no limitation on the dimension of the vector or the number of bits of the array.
  • the encoding result may also be a 64-bit array, and the terminal device 110 may encode 264 (2 to the 64th power) URLs, and the encoding results of different URLs may be different.
  • the first encoding manner may further include: index (index) encoding (dictionary-based encoding), multiple-hot (multiple-hot) encoding, etc., and the first encoding manner is not limited here.
  • the terminal device 110 uses the first hash algorithm to perform hash calculation on the encoding result corresponding to the historical behavior information of the first user in the second service, and obtains the hash corresponding to the historical behavior information of the first user in the second service value.
  • the terminal device 110 uses the first hash algorithm to perform hash calculation on the encoding result corresponding to the historical behavior information of the first user in the second service, if the similarity between the two encoding results is high, the first hash algorithm is used.
  • the Hash value after Hash algorithm performs hash calculation on the two encoding results also has a certain similarity. That is, by using the first hash algorithm to perform hash calculation on the encoding result, the similarity of the encoding result can be maintained.
  • the similarity of the encoding result can be maintained, so in the following S305, the user portrait server 130, according to the hash value corresponding to the user's historical behavior information in the second service, When a user performs user grouping, users with relatively similar historical behavior information can be divided into the same user group.
  • the first hash algorithm may be a locality sensitive hash (locality sensitive hashing, LSH) algorithm, or called an LSH function.
  • LSH locality sensitive hashing
  • the LSH algorithm is used to perform hash calculation on the encoding result, which can satisfy the similarity of the encoding result mentioned above.
  • the terminal device 110 adopts LSH
  • the step of the algorithm performing hash calculation on the encoding results corresponding to the URL A, the URL B, and the URL C may include: merging the encoding results corresponding to the URL A, the URL B, and the URL C respectively, to obtain the combined encoding result; Hash calculation is performed on the merged coding result by using the LSH algorithm to obtain a first hash value, which is the hash value corresponding to the three URLs of URL A, URL B, and URL C.
  • the encoding result corresponding to URL A is: [0, 1, 0, 0, 0, 0, 0]
  • the encoding result corresponding to URL B is: [0, 0, 1, 0, 0, 0, 0 , 0]
  • the encoding result corresponding to URL C is: [0, 0, 0, 0, 0, 1, 0, 0]
  • the terminal device 110 may use the LSH algorithm to perform hash calculation on [0, 1, 1, 0, 0, 1, 0, 0] to obtain the first hash value.
  • the first hash value is the hash value corresponding to the historical behavior information of the first user in the search service. This hash value may also be referred to as a hash fingerprint.
  • the LSH algorithm may adopt any of the following: the minimum hash (min-hash) algorithm using Jaccard to measure the similarity of data, and the algorithm using Euclidean distance to measure the similarity of data P-stable hash (P-stable hash) algorithm, simhash algorithm, etc., are not limited here.
  • the terminal device 110 sends the hash value corresponding to the historical behavior information of the first user in the second service to the user portrait server 130 .
  • the user portrait server 130 may receive a hash value corresponding to the historical behavior information of the first user in the second service.
  • the user portrait server 130 performs user grouping on the first user according to the hash value corresponding to the historical behavior information of the first user in the second service, and determines the user group to which the first user belongs in the second service.
  • the step of user grouping of the first user by the user portrait server 130 according to the hash value corresponding to the historical behavior information of the first user in the second service may include: the user portrait server 130 grouping the first user in the second service
  • the hash value corresponding to the historical behavior information in the second service, and the hash value corresponding to the historical behavior information of one or more users (such as the second user, the third user, etc.) in the second service are clustered and grouped , to obtain one or more user groups, and each user group may have a unique group identifier, such as a group ID.
  • the hash value corresponding to the historical behavior information of other users (such as the second user, the third user, etc.) in the second service can be sent to the user portrait server 130 by the terminal device of other users, and the terminal device 110 obtains the first
  • the principle of sending the hash value corresponding to the historical behavior information of the user in the second service to the user portrait server 130 is the same, and will not be repeated here.
  • the encoding result obtained by using the first encoding method to encode the historical behavior information related to the first user in the second service can be called the first encoding result.
  • An encoding result obtained by encoding the historical behavior information in the second service may be referred to as a second encoding result.
  • the user portrait server 130 may perform user grouping on the first user according to the hash value corresponding to the historical behavior information of the first user in the second service.
  • the similarity between the hash value corresponding to the historical behavior information of other users in the second service in some user groups and the hash value corresponding to the historical behavior information of the first user in the second service is greater than a certain threshold (eg, 90%, 80%, etc.), it is determined that the first user belongs to the existing user group.
  • a certain threshold eg, 90%, 80%, etc.
  • the existing user groups in the second service are obtained by clustering and grouping the hash values corresponding to the historical behavior information of other users (such as the second user, the third user, etc.) in the second service.
  • the other users in the aforementioned existing user group may be any user in the existing user group.
  • the hash values corresponding to the historical behavior information of other users in the aforementioned existing user group in the second service may be based on the respective hash values of a plurality of other users in the existing user group in the second service.
  • the hash value corresponding to the historical behavior information is calculated, such as: calculating the average value.
  • the user portrait server 130 sends to the terminal device 110 the group identification of the user group in which the first user belongs in the second service.
  • the group identifier may be used to indicate which user group the first user belongs to in the second service.
  • the terminal device 110 may receive and save the group identification of the user group in which the first user belongs in the second service.
  • the terminal device 110 When detecting that the first user uses the first service for the first time, the terminal device 110 sends to the recommendation server 120 a group identifier of a user group in which the first user belongs in the second service.
  • the recommendation server 120 may receive the group identification of the user group in which the first user belongs in the second service.
  • the first user may log in the user account of the first user on the terminal device 110 or an App installed on the terminal device 110 for providing the first service, and the terminal device 110 detects that the first user uses the first service for the first time.
  • a service means that the terminal device 110 detects that the user account of the first user opens or activates the first service.
  • the terminal device 110 may be a mobile phone, and a browser may be installed on the mobile phone, and the browser includes an information flow service, and the information flow service may be the aforementioned first service.
  • the first user can log in his user account on the browser, and when the mobile phone detects that the user account of the first user activates the information streaming service for the first time, it means that the first user uses the information streaming service for the first time.
  • the above-mentioned group identification of the user group in which the first user belongs in the second service may refer to: the group identification of the user group in which the user account of the first user belongs in the second service.
  • the recommendation server 120 sends to the terminal device 110, according to the group identifier of the user group in which the first user belongs to in the second service, the first user who is interested in other users who are in the same user group as the first user in the second service.
  • the service content of the service is not limited to the group identifier of the user group in which the first user belongs to in the second service.
  • the historical behavior of other users (such as the second user, the third user, etc.) in the second service
  • the hash value corresponding to the information is clustered and grouped to obtain one or more user groups, and the recommendation server 120 may record which service content of the first service the users corresponding to each user group are interested in. It should be noted that the recommendation server 120 may only record which service content of the first service the overall users corresponding to each user group are interested in, and may not pay attention to which specific user in the user group is interested in which service content of the first service interested.
  • the recommendation server 120 may, according to the group ID of the user group that the first user belongs to in the second service, query which service content of the first service the users corresponding to the user group corresponding to the group ID are interested in as a whole, that is, may Query which service content of the first service other users (such as the second user, the third user, etc.) included in the user group corresponding to the group identifier are interested in. Then, the recommendation server 120 may send to the terminal device 110 the service content of the first service that is of interest to other users in the same user group as the first user in the second service.
  • the service content of the first service that other users in the same user group as the first user in the second service are interested in may include: the service content of the first service browsed by other users, or, according to user
  • the portrait server 130 calculates and analyzes the interests of other users in the first service, and recommends the service content of the first service to other users.
  • the terminal device 110 displays the service content of the first service that other users in the same user group as the first user are interested in in the second service.
  • the terminal device 110 may recommend to the first user the service content of the first service that is of interest to other users in the same user group as the first user in the second service as a personalized recommendation result.
  • the terminal device 110 may use the information flow that is of interest to other users in the same user group as the first user in the search service as the information flow service.
  • the personalized recommendation results in are recommended to the first user.
  • the steps described in S301-S306 are the process of grouping users when they use the second service.
  • the steps described in S307-S309 are the process of cold-starting the user based on the user grouping results in S301-S306 when the user uses the first service for the first time.
  • the terminal device 110 may periodically execute the steps described in S301-S304 according to the first cycle.
  • the first period may be a day, a week, a month, etc.
  • the user portrait server 130 may periodically execute the steps described in S305-S306 along with the first cycle.
  • the historical behavior information of the first user in the second service acquired by the terminal device 110 may be the historical behavior information of the first user in the second service within a first time period.
  • the first time period may be the last month (ie), the last week, and so on.
  • the user grouping result of the first user can be dynamically updated according to the first period along with the change of the historical behavior information of the first user in the second service.
  • the user grouping results of the first user are dynamically updated according to the first cycle as the historical behavior information of the first user in the second service changes, which can not only improve the accuracy of the user grouping results, but also improve the security of user privacy .
  • the terminal device 110 can at least keep the history of the first user in the second service in the last month behavioral information. Every other week, the terminal device 110 can acquire the historical behavior information of the first user in the second service in the last month, and encode the historical behavior information of the first user in the second service in the last month using the first encoding method Processing, using the first hash algorithm to perform hash calculation on the encoding result corresponding to the historical behavior information of the first user in the second service in the last month, and obtain the historical behavior of the first user in the second service in the last month The hash value corresponding to the information is sent to the user portrait server 130.
  • the user portrait server 130 may perform the steps described above in S305-S306 based on the hash value corresponding to the historical behavior information of the first user in the second service within the last month, so that the terminal device 110 receives the first
  • the group identifier of the user group that a user belongs to in the second service can be updated every other week. It should be understood that after each update of the group identification of the user group that the first user belongs to in the second service, the results before and after the update may be the same or different, for example, the first user may be classified into a new user group Group.
  • the first user may also set on the terminal device 110 to choose to refuse to authorize the terminal device 110 to collect the historical behavior information of the first user in the service.
  • the terminal device 110 may no longer calculate the corresponding hash value according to the historical behavior information of the first user in the service.
  • the user portrait server 130 when the user portrait server 130 performs user grouping according to the hash value corresponding to the historical behavior information of the user (may include the first user, other users, etc.) in the second service, it may Keep the number of users in each user group below a fixed number, for example, keep the number of users in each user group less than 5000, so as to avoid the personalized recommendation results based on user grouping results being too single.
  • the step of encoding the historical behavior information of the first user in the second service by using the first encoding method is completed by the terminal device 110 .
  • the step of encoding the historical behavior information of the first user in the second service by using the first encoding method can also be completed on the side of the user portrait server 130, and the first user will be encoded by the first encoding method
  • the process of encoding the historical behavior information in the second service is placed on the user profile server 130 side, which can reduce the calculation pressure on the terminal device 110 side and reduce the load on the terminal device 110 .
  • FIG. 4 is another schematic flowchart of the recommendation method provided by the embodiment of the present application.
  • the recommendation method may include S401-S411.
  • the terminal device 110 acquires historical behavior information of a first user in a second service.
  • the terminal device 110 sends historical behavior information of the first user in the second service to the user portrait server 130.
  • the user portrait server 130 may receive historical behavior information of the first user in the second service.
  • the terminal device 110 may only send the historical behavior information of the first user in the second service to the user portrait server 130, and may not carry user information related to the first user, such as: user identification (user account), device information of the terminal device 110, etc.
  • the user portrait server 130 encodes the historical behavior information of the first user in the second service by using the first encoding method, and obtains an encoding result corresponding to the historical behavior information of the first user in the second service.
  • the process in which the user portrait server 130 uses the first encoding method to encode the historical behavior information of the first user in the second service can be compared with the terminal device 110 described in the previous embodiment using the first encoding method to encode the first user in the second service.
  • the process of encoding the historical behavior information in the second service is the same, and will not be repeated here.
  • the user portrait server 130 sends to the terminal device 110 an encoding result corresponding to the historical behavior information of the first user in the second service.
  • the terminal device 110 may receive an encoding result corresponding to the historical behavior information of the first user in the second service.
  • the terminal device 110 uses the first hash algorithm to perform hash calculation on the encoding result corresponding to the historical behavior information of the first user in the second service, and obtains the hash corresponding to the historical behavior information of the first user in the second service value.
  • the terminal device 110 sends the hash value corresponding to the historical behavior information of the first user in the second service to the user portrait server 130.
  • the user portrait server 130 may receive a hash value corresponding to the historical behavior information of the first user in the second service.
  • the user portrait server 130 performs user grouping on the first user according to the hash value corresponding to the historical behavior information of the first user in the second service, and determines the user group in which the first user belongs to in the second service.
  • the user portrait server 130 sends to the terminal device 110 the group identification of the user group in which the first user belongs in the second service.
  • the terminal device 110 may receive and save the group identification of the user group in which the first user belongs in the second service.
  • the terminal device 110 When detecting that the first user uses the first service for the first time, the terminal device 110 sends to the recommendation server 120 a group identifier of a user group in which the first user belongs in the second service.
  • the recommendation server 120 may receive the group identification of the user group in which the first user belongs in the second service.
  • the recommendation server 120 sends to the terminal device 110, according to the group identification of the user group in which the first user belongs to in the second service, the first user who is interested in other users who are in the same user group as the first user in the second service.
  • the service content of the service is not limited
  • the terminal device 110 displays service content of the first service that is of interest to other users in the same user group as the first user in the second service.
  • the functions of the terminal device 110 described in the embodiments of the present application may be realized by an App installed on the terminal device 110 .
  • a browser is installed on the terminal device 110, and the browser may include the above-mentioned first service and the second service, and all functions of the terminal device 110 described in the foregoing embodiments may be realized by the browser.
  • the terminal device 110 provides the first service and the second service
  • the first user uses the first service for the first time
  • the first user is a user who has used the second service as an example.
  • the recommended method provided by the embodiment of this application is introduced.
  • the recommendation method provided in the embodiments of the present application may also be applied in a multi-device coordination scenario, and the first service and the second service may be provided by two different devices.
  • multiple terminal devices such as mobile phones, tablet computers, personal computers (PCs), and smart home devices (such as televisions) can be used in coordination, and the scenario where multiple terminal devices cooperate can be called For multi-device collaboration scenarios.
  • the first user may have multiple devices capable of coordination, such as: terminal device 1, terminal device 2, and so on.
  • FIG. 5 is another schematic flowchart of the recommendation method provided by the embodiment of the present application.
  • the recommendation method may include S501-S511.
  • the terminal device 2 acquires historical behavior information of a first user in a second service.
  • the terminal device 2 sends historical behavior information of the first user in the second service to the user portrait server 130 .
  • the user portrait server 130 may receive historical behavior information of the first user in the second service.
  • the terminal device 2 may only send the historical behavior information of the first user in the second service to the user portrait server 130, and may not carry user information related to the first user, such as: user identification (user account), device information of the terminal device 2, etc.
  • the user portrait server 130 uses the first encoding method to encode the historical behavior information of the first user in the second service, and obtains the encoding result corresponding to the historical behavior information of the first user in the second service.
  • the process in which the user profile server 130 uses the first encoding method to encode the historical behavior information of the first user in the second service can be compared with the process of using the first encoding method to encode the first user in the second service described in the previous embodiment.
  • the process of encoding the historical behavior information in is the same and will not be repeated here.
  • the user portrait server 130 sends to the terminal device 2 an encoding result corresponding to the historical behavior information of the first user in the second service.
  • the terminal device 2 may receive an encoding result corresponding to the historical behavior information of the first user in the second service.
  • the process of encoding the historical behavior information of the first user in the second service by using the first encoding method may also be completed by the terminal device 2 . That is, the steps described in S502-S504 can also be replaced by: the terminal device 2 uses the first encoding method to encode the historical behavior information of the first user in the second service, and obtains the information of the first user in the second service. The encoding result corresponding to historical behavior information.
  • the terminal device 2 uses the first hash algorithm to perform hash calculation on the encoding result corresponding to the historical behavior information of the first user in the second service, and obtains the hash corresponding to the historical behavior information of the first user in the second service value.
  • the terminal device 2 sends the hash value corresponding to the historical behavior information of the first user in the second service to the user portrait server 130 .
  • the user portrait server 130 may receive a hash value corresponding to the historical behavior information of the first user in the second service.
  • the user portrait server 130 performs user grouping on the first user according to the hash value corresponding to the historical behavior information of the first user in the second service, and determines the user group to which the first user belongs in the second service.
  • the user portrait server 130 sends to the terminal device 1 the group identifier of the user group in which the first user belongs in the second service.
  • the terminal device 1 may receive and save the group identification of the user group in which the first user belongs in the second service.
  • the user portrait server 130 may send to the terminal device 2 the group ID of the user group that the first user belongs to in the second service, and the terminal device 1 may obtain from the terminal device 2 the user ID of the user group that the first user belongs to in the second service.
  • the grouping ID of the group may be used to obtain from the terminal device 2 the user ID of the user group that the first user belongs to in the second service.
  • the terminal device 1 When detecting that the first user uses the first service for the first time, the terminal device 1 sends to the recommendation server 120 a group identifier of a user group in which the first user belongs in the second service.
  • the recommendation server 120 may receive the group identification of the user group in which the first user belongs in the second service.
  • the recommendation server 120 sends to the terminal device 1, according to the group identifier of the user group in which the first user belongs to in the second service, the first user who is interested in other users who are in the same user group as the first user in the second service.
  • the service content of the service is not limited to the group identifier of the user group in which the first user belongs to in the second service.
  • the terminal device 1 displays service content of the first service that other users in the same user group as the first user are interested in in the second service.
  • the terminal device (which may be the same terminal device or different terminal devices) capable of providing the above-mentioned first service and second service in the embodiment of the present application may be a device resource (such as CPU resource, Memory resources, storage resources, etc.) are relatively rich, and can be independently networked and human-computer interaction devices, such as: mobile phones, tablet computers, notebook computers, etc.
  • a device resource such as CPU resource, Memory resources, storage resources, etc.
  • human-computer interaction devices such as: mobile phones, tablet computers, notebook computers, etc.
  • the terminal device that can provide the above-mentioned first service and/or second service in the embodiment of this application may also be a device that has less device resources and cannot be independently connected to the Internet and/or human-computer interaction, such as: handwriting Boards, headphones, VR glasses, car equipment, rice cookers, stereos, treadmills, etc.
  • the terminal device providing the first service is a device with less device resources and cannot independently network and/or human-computer interaction
  • the interaction between the terminal device providing the first service and the server (such as recommendation server, user portrait service) can be done through Other equipment resources are relatively rich, and can be completed by independent networking and human-computer interaction equipment.
  • the interaction between the terminal device providing the second service and the server (such as recommendation server, user portrait service) , It can also be completed by other devices with abundant resources, independent networking and human-computer interaction.
  • FIG. 6 is a schematic diagram of an application scenario provided by an embodiment of the present application.
  • a possible application scenario may include: mobile phone 601 , watch 602 , smart screen 603 , stereo 604 , refrigerator 605 and other devices.
  • the mobile phone 601 and the watch 602 , smart screen 603 , stereo 604 , and refrigerator 605 can be connected via a router 606 via a local area network.
  • mobile phone 601, watch 602, smart screen 603, etc. can be considered as devices with relatively rich equipment resources, which can be independently connected to the Internet and interact with humans; audio equipment 604, refrigerator 605, etc. can be considered as equipment with relatively few resources. Devices that cannot be independently networked and/or human-computer interactive.
  • the first service may be a recipe recommendation service provided by the refrigerator 605
  • the second service may be an information streaming service provided by the mobile phone 601 .
  • the refrigerator 605 may be the terminal device 1 mentioned in the process shown in FIG. 5 above
  • the mobile phone 601 may be the terminal device 2 mentioned in the process shown in FIG. 5 above.
  • the refrigerator 605 can complete the interaction with the server (such as recommendation server, user portrait service) through devices such as mobile phone 601, watch 602, or smart screen 603 that have rich resources and can be independently networked and human-computer interaction.
  • the server such as recommendation server, user portrait service
  • the user portrait server may send the group identifier of the user group that the first user belongs to in the second service to the mobile phone 601, and the mobile phone 601 may forward the group identifier of the user group that the first user belongs to in the second service to the refrigerator 605; the refrigerator 605 may send to the mobile phone 601 the group identifier of the user group in which the first user belongs in the second service, and the mobile phone 601 may forward the group identifier of the user group in which the first user belongs in the second service to the recommendation server
  • the recommendation server can send to the mobile phone 601 the service content of the first service that other users who are in the same user group as the first user in the second service are interested in, and the mobile phone 601 can be in the same user group as the first user in the second service.
  • the service content of the first service that other users of the user group are interested in is forwarded to the refrigerator 605 .
  • the first service may be a recipe recommendation service provided by the refrigerator 605
  • the second service may be a music recommendation service provided by the stereo 604
  • the refrigerator 605 may be the terminal device 1 mentioned above in the process shown in FIG. 5
  • the stereo 604 may be the terminal device 2 mentioned in the process shown in FIG. 5 above.
  • the refrigerator 605 can complete the interaction with the server (such as recommendation server, user portrait service) through devices such as mobile phone 601, watch 602, or smart screen 603 that have rich resources and can be independently networked and human-computer interaction.
  • the audio 604 can also interact with the server (such as the recommendation server, user portrait service) through devices such as the mobile phone 601, or the watch 602, or the smart screen 603, which are relatively rich in resources and can be independently networked and human-computer interaction.
  • the server such as recommendation server, user portrait service
  • the above-mentioned devices with rich device resources, independent networking and human-computer interaction can be called rich devices, and the above-mentioned devices with few device resources, which cannot be independently connected to the Internet and/or human-computer interaction can be called rich devices.
  • rich devices For light equipment.
  • the recommendation method provided in the embodiment of the present application may also be applied to a non-user cold start scenario.
  • the first user may not be a new user in the first service, but a user who has used the first service. That is, the first user has generated some historical behavior information in the first service.
  • the recommendation method provided by the embodiment of the present application can also use the historical behavior information related to the first user in the second service to carry out knowledge transfer, determine the user grouping results of the first user in the second service, and based on the second service The user grouping results of a user in the second service are collaboratively recommended to the user in the first service.
  • the first user may be a user who has used the first service, but the historical behavior information of the first user in the first service is less. This application does not limit whether the first user is using the first service for the first time, and the amount of historical behavior information of the first user in the first service.
  • the terminal device 110 may be referred to as a first device.
  • terminal device 1 may be called a first device
  • terminal device 2 may be called a second device.
  • the other terminal devices may be called third devices.
  • the first device may be the aforementioned light device
  • the third device may be the aforementioned rich device.
  • the second device interacts with the server through other terminal devices
  • the other terminal devices may be referred to as fourth devices.
  • the first device may be the aforementioned light device
  • the fourth device may be the aforementioned rich device.
  • the first device may also send identification information of the first user (such as a user account) to the server.
  • the first device may send user information of the first user to the server, where the user information of the first user includes: identification information of the first user and identification information of the first user group.
  • the first device may not perceive the group identifier of the user group that the first user belongs to in the second service, and the server obtains the user group identifier of the user group that the first user belongs to in the second service After the group identifier of the user group, an association relationship between the group identifier of the user group that the first user belongs to in the second service and the first user's identification information (such as a user account) can be established on the server side.
  • the first device can send the identification information of the first user to the server, and the server can obtain the group identification of the user group that the first user belongs to in the second service according to the identification information of the first user.
  • the identification information query obtains which user group the first user belongs to in the second service.
  • the user group that the first user belongs to in the second service may be referred to as the first user group
  • the group identifier of the user group that the first user belongs to in the second service is the ID of the first user group.
  • Identification information That is, the first device may send the user information of the first user to the server, and the user information of the first user may include: identification information of the first user, and/or identification information of the first user group. Wherein, the user information of the first user is related to the first user group.
  • the server can use the hash value corresponding to the historical behavior information related to the first user in the second service and the history of one or more other users in the second service Matching the hash value corresponding to the behavior information, determining the user group that the first user belongs to in the second service (that is, the above-mentioned first user group) is described as an example, but in some other possible embodiments, the hash value The Greek value can also be replaced by the result obtained by using other feature extraction algorithms to extract features from the coding results corresponding to the historical behavior information, which is not limited in this application.
  • the results obtained by other feature extraction algorithms after performing feature extraction on the encoding results corresponding to historical behavior information can also maintain the similarity of the encoding results, and other feature extraction algorithms can also have the characteristic that they cannot reversely decrypt the encoding results.
  • the hash value corresponding to the aforementioned historical behavior information, the result obtained after using other feature extraction algorithms to perform feature extraction on the encoding result corresponding to the historical behavior information, etc. can be called the feature information corresponding to the historical behavior information. That is, in this application, it can be matched according to the feature information corresponding to the historical behavior information related to the first user in the second service and the feature information corresponding to the historical behavior information of one or more other users in the second service, A user group to which the first user belongs in the second service is determined.
  • an embodiment of the present application provides a recommendation system, and the recommendation system can implement the recommendation method as described in the foregoing embodiments.
  • the recommendation system may include: a first device and a server.
  • the first device may be configured to send user information of the first user to the server.
  • the server can be used to send recommendation information to the first device.
  • the recommended information is information related to the use of the first service by users in the first user group; the user information of the first user is related to the first user group; the first user group is based on the information related to the first user in the second service.
  • the characteristic information corresponding to the historical behavior information of the user and the characteristic information corresponding to the historical behavior information of the other one or more users in the second service are determined by matching.
  • the first device includes a first service and a second service; the first device is further configured to send feature information corresponding to historical behavior information related to the first user in the second service to the server.
  • the recommendation system further includes a second device.
  • the first device includes a first service and the second device includes a second service.
  • the second device may be configured to send feature information corresponding to historical behavior information related to the first user in the second service to the server.
  • the recommendation system further includes a third device, and the first device is specifically configured to send the user information of the first user to the server through the third device.
  • the server is specifically configured to send recommendation information to the first device through the third device.
  • the user information of the first user includes: identification information of the first user, and/or identification information of the first user group.
  • the server is further configured to determine the second service according to the characteristic information corresponding to the historical behavior information related to the first user in the second service and the characteristic information corresponding to the historical behavior information of one or more other users in the second service.
  • a user group and, sending identification information of the first user group to the first device.
  • the characteristic information corresponding to the historical behavior information related to the first user in the second service includes: a hash value corresponding to the historical behavior information related to the first user in the second service.
  • the feature information corresponding to the historical behavior information of the other one or more users in the second service includes: a hash value corresponding to the historical behavior information of the other one or more users in the second service.
  • the hash value corresponding to the historical behavior information related to the first user in the second service is obtained by hashing the first encoding result using the first hash algorithm, and the first encoding result is obtained by using the first encoding
  • the method is obtained by encoding the historical behavior information related to the first user in the second service.
  • the hash value corresponding to the historical behavior information of one or more users in the second service is obtained by hashing the second encoding result using the first hash algorithm, and the second encoding result is obtained by using the first encoding method to It is obtained by encoding the historical behavior information of one or more other users in the second service.
  • the first device is specifically configured to, when detecting that the first user uses the first service for the first time, send the user information of the first user to the server.
  • the recommendation system can implement the functions corresponding to all the steps of the recommendation method described in the foregoing embodiments, which will not be repeated here.
  • An embodiment of the present application also provides a recommendation apparatus, which can be applied to an electronic device (such as a first device), so that the electronic device implements the steps performed by the first device in the recommendation method described in the foregoing embodiments.
  • the functions of the device can be realized by hardware, and can also be realized by executing corresponding software by hardware.
  • the hardware or software includes one or more modules or units corresponding to the steps performed by the first device in the recommended method described in the foregoing embodiments.
  • FIG. 7 is a schematic structural diagram of a recommendation device provided in an embodiment of the present application. As shown in FIG. 7, the apparatus may include: a sending unit 701, a receiving unit 702, and the like.
  • the sending unit 701 may be used to send user information of the first user to the server; the receiving unit 702 may be used to receive recommendation information from the server; the recommendation information is information related to the use of the first service by users in the first user group; the second The user information of a user is related to the first user group; the first user group is based on the feature information corresponding to the historical behavior information related to the first user in the second service, and other one or more users in the second service The feature information corresponding to the historical behavior information in the service is determined by matching.
  • the first device includes a first service and a second service; the sending unit 701 is further configured to send feature information corresponding to historical behavior information related to the first user in the second service to the server.
  • the first device includes the first service
  • the second device includes the second service
  • the sending unit 701 is specifically configured to send the user information of the first user to the server through the third device.
  • the receiving unit 702 is specifically configured to receive recommendation information from the server through the third device.
  • the sending unit 701 is specifically configured to send the feature information corresponding to the historical behavior information related to the first user in the second service to the server through the third device.
  • the receiving unit 702 is also configured to receive the identification information of the first user group sent from the server; the first user group is the characteristic information corresponding to the historical behavior information related to the first user in the second service by the server , and other feature information corresponding to the historical behavior information of one or more users in the second service.
  • the sending unit 701 is specifically configured to send the user information of the first user to the server when it is detected that the first user uses the first service for the first time.
  • the apparatus can implement the functions corresponding to all the steps performed by the first device in the recommendation method described in the foregoing embodiments, which will not be repeated here.
  • An embodiment of the present application also provides a recommendation apparatus, which can be applied to an electronic device (such as a server), so that the electronic device implements the steps performed by the server in the recommendation method described in the foregoing embodiments.
  • the functions of the device can be realized by hardware, and can also be realized by executing corresponding software by hardware.
  • the hardware or software includes one or more modules or units corresponding to the steps executed by the server in the recommended method described in the foregoing embodiments.
  • FIG. 8 is another schematic structural diagram of a recommendation device provided in an embodiment of the present application. As shown in FIG. 8 , the apparatus may include: a receiving unit 801 and a sending unit 802 .
  • the receiving unit 801 may be configured to receive user information of the first user sent from the first device; the sending unit 802 may be configured to send recommendation information to the first device.
  • the recommended information is information related to the use of the first service by users in the first user group; the user information of the first user is related to the first user group; the first user group is based on the information related to the first user in the second service.
  • the characteristic information corresponding to the historical behavior information of the user and the characteristic information corresponding to the historical behavior information of the other one or more users in the second service are determined by matching.
  • the first device includes the first service and the second service; the receiving unit 801 is further configured to receive the characteristic information corresponding to the historical behavior information related to the first user in the second service sent by the first device. information.
  • the first device includes the first service
  • the second device includes the second service
  • the receiving unit 801 is further configured to receive the history related to the first user in the second service sent from the second device The characteristic information corresponding to the behavior information.
  • the receiving unit 801 is specifically configured to receive, through the third device, the user information of the first user sent from the first device.
  • the sending unit 802 is specifically configured to send recommendation information to the first device through the third device.
  • the receiving unit 801 is specifically configured to receive, through the third device, characteristic information corresponding to historical behavior information related to the first user in the second service sent by the first device.
  • the receiving unit 801 is specifically configured to receive, through the fourth device, feature information corresponding to historical behavior information related to the first user in the second service sent by the second device.
  • the apparatus may further include: a processing unit 803 .
  • the processing unit 803 may be configured to determine the first user according to the feature information corresponding to the historical behavior information related to the first user in the second service and the feature information corresponding to the historical behavior information of one or more other users in the second service. group.
  • the sending unit 802 is further configured to send the identification information of the first user group to the first device.
  • the device can implement the functions corresponding to all the steps performed by the server in the recommendation method described in the foregoing embodiments, which will not be repeated here.
  • the division of units (or called modules) in the above device is only a division of logical functions, and may be fully or partially integrated into a physical entity or physically separated during actual implementation.
  • the units in the device can all be implemented in the form of software called by the processing element; they can also be implemented in the form of hardware; some units can also be implemented in the form of software called by the processing element, and some units can be implemented in the form of hardware.
  • each unit can be a separate processing element, or it can be integrated in a certain chip of the device. In addition, it can also be stored in the memory in the form of a program, which is called and executed by a certain processing element of the device. Function. In addition, all or part of these units can be integrated together, or implemented independently.
  • the processing element described here may also be referred to as a processor, and may be an integrated circuit with a signal processing capability. In the process of implementation, each step of the above method or each unit above may be implemented by an integrated logic circuit of hardware in the processor element or implemented in the form of software called by the processing element.
  • the units in the above device may be one or more integrated circuits configured to implement the above method, for example: one or more application specific integrated circuits (ASIC), or, one or more A digital signal processor (DSP), or, one or more field programmable gate arrays (FPGA), or a combination of at least two of these integrated circuit forms.
  • ASIC application specific integrated circuits
  • DSP digital signal processor
  • FPGA field programmable gate arrays
  • the processing element can be a general-purpose processor, such as a central processing unit (central processing unit, CPU) or other processors that can call programs.
  • CPU central processing unit
  • these units can be integrated together and implemented in the form of a system-on-a-chip (SOC).
  • the units of the above apparatus for implementing each corresponding step in the above method may be implemented in the form of a processing element scheduler.
  • the apparatus may include a processing element and a storage element, and the processing element invokes a program stored in the storage element to execute the steps performed by the first device or the server in the recommended method described in the above method embodiments.
  • the storage element may be a storage element on the same chip as the processing element, that is, an on-chip storage element.
  • the program for executing the above method may be stored in a storage element on a different chip from the processing element, that is, an off-chip storage element.
  • the processing element invokes or loads a program from the off-chip storage element on the on-chip storage element, so as to invoke and execute the steps performed by the first device or the server in the recommended method described in the above method embodiments.
  • the embodiment of the present application also provides an electronic device.
  • the electronic device may be the above-mentioned first device or server.
  • the electronic device includes: a processor, a memory for storing processor-executable instructions; when the processor is configured to execute the instructions, the electronic device implements the first device or the server in the recommended method described in the above method embodiments.
  • the memory can be located inside the electronic device or outside the electronic device.
  • the processor includes one or more.
  • the unit of the electronic device that implements each step in the above method may be configured as one or more processing elements, where the processing elements may be integrated circuits, for example: one or more ASICs, or one Or multiple DSPs, or, one or more FPGAs, or a combination of these types of integrated circuits. These integrated circuits can be integrated together to form a chip.
  • an embodiment of the present application further provides a chip, and the chip can be applied to the above-mentioned electronic device.
  • the chip includes one or more interface circuits and one or more processors; the interface circuits and processors are interconnected through lines; the processor receives and executes computer instructions from the memory of the electronic device through the interface circuits, so as to realize the above-mentioned method embodiments. The steps performed by the first device or server in the recommended method.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, each unit may exist separately physically, or two or more units may be integrated into one unit.
  • the above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.
  • the integrated unit is realized in the form of a software function unit and sold or used as an independent product, it can be stored in a readable storage medium.
  • the software product is stored in a program product, such as a computer-readable storage medium, and includes several instructions to make a device (which may be a single-chip microcomputer, a chip, etc.) or a processor (processor) execute all of the methods described in various embodiments of the present application. or partial steps.
  • the aforementioned storage medium includes: various media capable of storing program codes such as U disk, mobile hard disk, ROM, RAM, magnetic disk or optical disk.
  • an embodiment of the present application also provides a computer-readable storage medium, on which computer program instructions are stored; when the computer program instructions are executed by an electronic device, the electronic device implements the recommended method as described in the above method embodiments Steps performed by the first device or the server.
  • an embodiment of the present application further provides a computer program product, including: computer readable codes, or a non-volatile computer readable storage medium bearing computer readable codes, when the computer readable codes are stored in the electronic device
  • the processor in the electronic device implements the steps performed by the first device or the server in the recommended method described in the above method embodiments.

Landscapes

  • Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Data Mining & Analysis (AREA)
  • Strategic Management (AREA)
  • General Engineering & Computer Science (AREA)
  • Development Economics (AREA)
  • Multimedia (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Game Theory and Decision Science (AREA)
  • Signal Processing (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Information Transfer Between Computers (AREA)

Abstract

La présente invention se rapporte au domaine du traitement de données, et concerne un procédé de recommandation, un dispositif électronique et un support de stockage. Selon le procédé, les informations de caractéristique dans un second service correspondant à des informations de comportement historiques relatives à un premier utilisateur, et les informations de caractéristique correspondant à des informations de comportement historiques du ou des autres utilisateurs dans le second service peuvent être mises en correspondance, un résultat de regroupement d'utilisateurs du premier utilisateur dans le second service peut être déterminé, et une recommandation personnalisée peut être effectuée pour le premier utilisateur dans un premier service sur la base du résultat de regroupement d'utilisateurs du premier utilisateur dans le second service. Grâce au procédé, les résultats de regroupement d'utilisateurs peuvent être réduits, la précision de la recommandation personnalisée peut être améliorée de manière efficace, et les résultats de la recommandation qui intéressent davantage le premier utilisateur peuvent être recommandés au premier utilisateur.
PCT/CN2022/112106 2021-10-29 2022-08-12 Procédé de recommandation, dispositif électronique et support de stockage WO2023071404A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202111276088.6A CN116070007A (zh) 2021-10-29 2021-10-29 推荐方法、电子设备及存储介质
CN202111276088.6 2021-10-29

Publications (1)

Publication Number Publication Date
WO2023071404A1 true WO2023071404A1 (fr) 2023-05-04

Family

ID=86159036

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/112106 WO2023071404A1 (fr) 2021-10-29 2022-08-12 Procédé de recommandation, dispositif électronique et support de stockage

Country Status (2)

Country Link
CN (1) CN116070007A (fr)
WO (1) WO2023071404A1 (fr)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116894692A (zh) * 2023-09-11 2023-10-17 北京亿家老小科技有限公司 一种线上网络销售用户潜在需求分析监控方法及系统
CN117235321A (zh) * 2023-09-04 2023-12-15 之江实验室 一种展览点位推荐方法、装置、电子装置和存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102044009A (zh) * 2009-10-23 2011-05-04 华为技术有限公司 群组推荐方法和系统
US20160321735A1 (en) * 2013-10-28 2016-11-03 Telefonaktiebolaget L M Ericsson (Publ) Group recommendation for selection of service items
CN107360246A (zh) * 2017-07-28 2017-11-17 广州优视网络科技有限公司 一种消息推送方法及装置、一种终端及存储介质
WO2018043860A1 (fr) * 2016-09-02 2018-03-08 에스케이플래닛 주식회사 Dispositif de recommandation d'article de location au moyen d'un groupe de propension similaire et procédé d'utilisation associé
CN109033443A (zh) * 2018-08-17 2018-12-18 腾讯科技（深圳）有限公司 推荐对象确定方法、装置、存储介质及计算机设备

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102044009A (zh) * 2009-10-23 2011-05-04 华为技术有限公司 群组推荐方法和系统
US20160321735A1 (en) * 2013-10-28 2016-11-03 Telefonaktiebolaget L M Ericsson (Publ) Group recommendation for selection of service items
WO2018043860A1 (fr) * 2016-09-02 2018-03-08 에스케이플래닛 주식회사 Dispositif de recommandation d'article de location au moyen d'un groupe de propension similaire et procédé d'utilisation associé
CN107360246A (zh) * 2017-07-28 2017-11-17 广州优视网络科技有限公司 一种消息推送方法及装置、一种终端及存储介质
CN109033443A (zh) * 2018-08-17 2018-12-18 腾讯科技（深圳）有限公司 推荐对象确定方法、装置、存储介质及计算机设备

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117235321A (zh) * 2023-09-04 2023-12-15 之江实验室 一种展览点位推荐方法、装置、电子装置和存储介质
CN117235321B (zh) * 2023-09-04 2024-04-16 之江实验室 一种展览点位推荐方法、装置、电子装置和存储介质
CN116894692A (zh) * 2023-09-11 2023-10-17 北京亿家老小科技有限公司 一种线上网络销售用户潜在需求分析监控方法及系统
CN116894692B (zh) * 2023-09-11 2023-11-24 北京亿家老小科技有限公司 一种线上网络销售用户潜在需求分析监控方法及系统

Also Published As

Publication number Publication date
CN116070007A (zh) 2023-05-05

Similar Documents

Publication Publication Date Title
WO2023071404A1 (fr) Procédé de recommandation, dispositif électronique et support de stockage
US20100299338A1 (en) Method and apparatus for providing device compatibility service
US10346737B1 (en) Distributed multisensor system to record spatially diverse events
WO2020220180A1 (fr) Procédé et dispositif de recommandation de contenu multimédia
WO2022160993A1 (fr) Procédé et dispositif de partage de données multimédias
WO2023273543A1 (fr) Procédé et appareil de gestion de dossier
WO2020062014A1 (fr) Procédé d'entrée d'informations dans une boîte d'entrée et dispositif électronique
WO2016173404A1 (fr) Procédé et dispositif de correspondance d'utilisateurs
WO2023010705A1 (fr) Procédé de traitement de données, terminal mobile et support de stockage
CN112148962B (zh) 用于推送信息的方法和装置
US20220311700A1 (en) Method for multiplexing http channels and terminal
WO2020107168A1 (fr) Procédé et dispositif de décodage vidéo, appareil électronique, et support de stockage lisible par ordinateur
US20230385345A1 (en) Content recommendation method, electronic device, and server
WO2012171195A1 (fr) Procédé et appareil pour rechercher un contenu dans un canal en fonction de caractéristiques contextuelles
JP2014501952A (ja) マルチキャストネットワークにわたる送信のためのコンテンツのプログラミングに影響を与えるための方法および装置
CN106034065A (zh) 信息展示方法及装置
CN110389805B (zh) 一种信息展示方法、装置、设备及存储介质
US20150302059A1 (en) Content recommendation apparatus and the method thereof
WO2019096207A1 (fr) Procédé de traitement d'image, dispositif informatique et support d'informations lisible par ordinateur
WO2019127654A1 (fr) Procédé et système d'identification de vidéos dangereuses sur la base d'un ip d'utilisateur et d'un contenu de crédits
CN110781066A (zh) 用户行为分析方法、装置、设备及存储介质
WO2019127659A1 (fr) Procédé et système permettant d'identifier une vidéo nocive sur la base d'un id d'utilisateur
WO2019127656A1 (fr) Procédé et système d'identification de vidéo dangereuse basés sur une copie vidéo et un ip d'utilisateur
WO2023001084A1 (fr) Procédé et appareil d'affichage d'image, procédé et appareil de traitement d'image, et dispositif
WO2019127651A1 (fr) Procédé et système associé permettant d'identifier une vidéo malveillante

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22885304

Country of ref document: EP

Kind code of ref document: A1