CN107436914B - Recommendation method and device - Google Patents

Recommendation method and device Download PDF

Info

Publication number
CN107436914B
CN107436914B CN201710419957.3A CN201710419957A CN107436914B CN 107436914 B CN107436914 B CN 107436914B CN 201710419957 A CN201710419957 A CN 201710419957A CN 107436914 B CN107436914 B CN 107436914B
Authority
CN
China
Prior art keywords
social
user
recommended
users
merchant
Prior art date
Legal status (The legal status 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 status listed.)
Active
Application number
CN201710419957.3A
Other languages
Chinese (zh)
Other versions
CN107436914A (en
Inventor
程亮
李泽中
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Xingxuan Technology Co Ltd
Original Assignee
Beijing Xingxuan Technology Co Ltd
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 Beijing Xingxuan Technology Co Ltd filed Critical Beijing Xingxuan Technology Co Ltd
Priority to CN201710419957.3A priority Critical patent/CN107436914B/en
Publication of CN107436914A publication Critical patent/CN107436914A/en
Application granted granted Critical
Publication of CN107436914B publication Critical patent/CN107436914B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • G06Q30/0241—Advertisements
    • G06Q30/0251—Targeted advertisements
    • G06Q30/0255—Targeted advertisements based on user history
    • 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
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/01—Social networking

Abstract

The embodiment of the application provides a recommendation method and device. The recommendation method comprises the following steps: determining a social user set corresponding to a user to be recommended; calculating the intimacy between the user to be recommended and each social user in the social user set; and recommending the interested commercial tenants to the user to be recommended from the commercial tenant set associated with the social user set according to the intimacy between the user to be recommended and each social user in the social user set. The embodiment of the application can improve the participation degree of the user for the recommended merchants and is beneficial to more fully playing the recommendation effect.

Description

Recommendation method and device
Technical Field
The application relates to the technical field of internet, in particular to a recommendation method and device.
Background
With the continuous maturity of recommendation technologies, personalized recommendation services are increasingly applied to the internet industry. The service quality of the application can be improved based on the personalized recommendation service, the number of users is increased for the application, and the like.
The recommendation method based on collaborative filtering is a common recommendation method, and the principle is as follows: calculating the similarity between other commodities and the commodities recently consumed by the user according to the commodities recently consumed by the user; and acquiring the commodities with higher similarity and recommending the commodities to the user.
The recommendation method based on collaborative filtering is not ideal enough in recommendation effect, the user participation based on the recommendation result is relatively low, and the recommendation effect is to be further improved.
Disclosure of Invention
The inventor of the application conducts tracking research on the using process of the existing recommendation method, finds that the existing recommendation method can improve the service quality of application to a certain extent, is beneficial to increasing the number of users applying the recommendation method, but is not ideal, and the user participation degree based on the recommendation result is low so that the recommendation effect is to be further improved.
Based on the above, the inventor of the present application provides a solution after creative work, and the main principle is: according to the intimacy between the social users, merchant recommendation is carried out from the associated merchants of the social users, so that the reliability of the recommended merchants is improved, the participation of the users in the recommended merchants is improved, and the recommendation effect is improved.
The embodiment of the application provides a recommendation method, which comprises the following steps:
determining a social user set corresponding to a user to be recommended;
calculating the intimacy between the user to be recommended and each social user in the social user set;
and recommending interest merchants to the user to be recommended from a merchant set associated with the social user set according to the intimacy between the user to be recommended and each social user in the social user set.
In an optional embodiment, determining a social user set corresponding to a user to be recommended includes:
capturing social data of the user to be recommended on a social contact App, a social contact website and/or an application App;
and mining users having social relations with the users to be recommended from the social data to form the social user set.
In an optional embodiment, crawling social data of the user to be recommended on a social App, a social website and/or an application App includes:
capturing relevant data of at least one social behavior of the user to be recommended on a social App, a social website and/or an application App:
the data comprises data related to chat behaviors, data related to sharing behaviors, data related to picking behaviors, data related to viewing behaviors, data related to commenting behaviors and data related to praise behaviors.
In an optional embodiment, mining, from the social data, users who have a social relationship with the user to be recommended to form the social user set, including:
taking the users in the social data as vertexes and taking the social relations among the users in the social data as edges to construct a social network graph of the user to be recommended;
and acquiring the users with social relations with the users to be recommended based on the social network diagram to form the social user set.
In an optional embodiment, calculating the affinity between the user to be recommended and each social user in the set of social users includes:
and calculating the intimacy between the user to be recommended and each social user in the social user set according to the attribute information of the social relationship existing between the user to be recommended and each social user in the social user set.
In an optional implementation manner, calculating the affinity between the user to be recommended and each social user in the set of social users according to the attribute information of the social relationship existing between the user to be recommended and each social user in the set of social users includes:
and calculating the intimacy between the user to be recommended and each social user in the social user set according to the type and/or the frequency of the social relationship between the user to be recommended and each social user in the social user set.
In an optional embodiment, calculating the affinity between the user to be recommended and each social user in the set of social users according to the number of times of the social relationship existing between the user to be recommended and each social user in the set of social users includes:
according to the formula
Figure BDA0001314690860000031
Calculating the intimacy between the user to be recommended and each social user in the social user set;
u0representing the user to be recommended;
Friends(u0) Representing the set of social users;
ujrepresenting the jth social user in the set of social users, 1<＝j<N, N represents the number of social users in the set of social users;
ukrepresenting the kth social user of the set of social users, 1<＝k<＝N；
Interactions(u0，uj) Representing the user u to be recommended0With social users ujThe number of social relationships that exist between;
Interactions(u0，uk) Representing the user u to be recommended0With social users ukThe number of social relationships that exist between;
Intimacy(u0，uj) Representing the user u to be recommended0With social users ujThe degree of intimacy therebetween.
In an optional embodiment, calculating the affinity between the user to be recommended and each social user in the set of social users according to the type and the number of social relationships includes:
calculating the child intimacy degree of each social relationship between the user to be recommended and each social user in the social user set according to the type and the frequency of the social relationship between the user to be recommended and each social user in the social user set;
and calculating the intimacy between the user to be recommended and each social user in the social user set according to the child intimacy of the user to be recommended and each social user in the social user set under each social relationship.
In an optional embodiment, recommending an interest business to the user to be recommended from a business set associated with the social user set according to the affinity between the user to be recommended and each social user in the social user set includes:
calculating the intimacy between each social user in the social user set and each merchant in the merchant set;
determining recommendation scores of all merchants in the merchant set according to the intimacy between the user to be recommended and all social users in the social user set and the intimacy between all social users in the social user set and all merchants in the merchant set;
and recommending the interested commercial tenant to the user to be recommended from the commercial tenant set according to the recommendation score of each commercial tenant in the commercial tenant set.
In an optional embodiment, calculating the affinity between each social user in the set of social users and each merchant in the set of merchants includes:
and calculating the intimacy between each social user in the social user set and each merchant in the merchant set according to the attribute information of the network behavior relationship between each social user in the social user set and each merchant in the merchant set.
In an optional embodiment, calculating, according to attribute information of a network behavior relationship existing between each social user in the set of social users and each merchant in the set of merchants, an affinity between each social user in the set of social users and each merchant in the set of merchants includes:
and calculating the intimacy between each social user in the social user set and each merchant in the merchant set according to the type and/or the times of the network behavior relationship existing between each social user in the social user set and each merchant in the merchant set.
In an optional embodiment, calculating the affinity between each social user in the set of social users and each merchant in the set of merchants according to the number of times of the network behavior relationship between each social user in the set of social users and each merchant in the set of merchants includes:
according to the formula
Figure BDA0001314690860000041
Calculating the intimacy between each social user in the social user set and each merchant in the merchant set;
ujrepresenting the jth social user in the set of social users, 1<＝j<N, N represents the number of social users in the set of social users;
sirepresenting the i-th merchant in the set of merchants, 1<＝i<M, representing the number of merchants in the set of merchants;
OrderShops(uj) Representing social users ujA set of merchants having a network behavioral relationship;
slrepresenting social users ujThe ith merchant in the set of merchants with network behavior relationship, 1<＝l<L denotes a social user ujThe number of merchants with network behavior relationships;
OrderTimes(ui，sl) Representing social users ujWith merchants slThe number of times there is a network behavior relationship between them;
OrderTimes(uj，si) Representing social users ujWith merchants siThe number of times there is a network behavior relationship between them;
Intimacy(uj，si) Representing social users ujWith merchants siThe degree of intimacy therebetween.
In an optional embodiment, determining a recommendation score of each merchant in the merchant set according to the affinity between the user to be recommended and each social user in the social user set and the affinity between each social user in the social user set and each merchant in the merchant set includes:
calculating the recommendation score of each merchant in the merchant set according to the following formula;
Figure BDA0001314690860000051
u0representing the user to be recommended;
Friends(u0) Representing the set of social users;
ujrepresenting the jth social user in the set of social users, 1<＝j<N, N represents the number of social users in the set of social users;
skrepresents the kth merchant in the set of merchants, 1<＝k<M, representing the number of merchants in the set of merchants;
Intimacy(u0，uj) Representing the user to be recommended and the social user uj(ii) the intimacy;
Intimacy(uj，sk) Representing social users ujBusiness with businessFamily sk(ii) the intimacy;
RecScore(sk) Representing said merchant skThe recommendation score of (1).
In an optional embodiment, before calculating the affinity between each social user in the set of social users and each merchant in the set of merchants, the method further includes:
and determining the merchant set from merchants having network behavior relations with any social users in the social user set.
An embodiment of the present application further provides a recommendation device, including:
the determining module is used for determining a social user set corresponding to a user to be recommended;
the calculation module is used for calculating the intimacy between the user to be recommended and each social user in the social user set;
and the recommending module is used for recommending the interested commercial tenant to the user to be recommended from the commercial tenant set associated with the social user set according to the intimacy between the user to be recommended and each social user in the social user set.
In an optional embodiment, the determining module comprises:
the crawling sub-module is used for crawling social data of the user to be recommended on the social App, the social website and/or the application App;
and the mining submodule is used for mining the users with social relations with the users to be recommended from the social data to form the social user set.
In an optional embodiment, the grasping sub-module is specifically configured to:
capturing relevant data of at least one social behavior of the user to be recommended on a social App, a social website and/or an application App:
the data comprises data related to chat behaviors, data related to sharing behaviors, data related to picking behaviors, data related to viewing behaviors, data related to commenting behaviors and data related to praise behaviors.
In an optional embodiment, the mining submodule is specifically configured to:
taking the users in the social data as vertexes and taking the social relations among the users in the social data as edges to construct a social network graph of the user to be recommended;
and acquiring the users with social relations with the users to be recommended based on the social network diagram to form the social user set.
In an optional embodiment, the calculation module is specifically configured to:
and calculating the intimacy between the user to be recommended and each social user in the social user set according to the attribute information of the social relationship existing between the user to be recommended and each social user in the social user set.
In an optional embodiment, the calculation module is specifically configured to:
and calculating the intimacy between the user to be recommended and each social user in the social user set according to the type and/or the frequency of the social relationship between the user to be recommended and each social user in the social user set.
In an optional embodiment, the calculation module is specifically configured to:
according to the formula
Figure BDA0001314690860000071
Calculating the intimacy between the user to be recommended and each social user in the social user set;
u0representing the user to be recommended;
Friends(u0) Representing the set of social users;
ujrepresenting the jth social user in the set of social users, 1<＝j<N, N represents the number of social users in the set of social users;
ukrepresenting the kth social user of the set of social users, 1<＝k<＝N；
Interactions(u0，uj) Representing the user u to be recommended0With social users ujThe number of social relationships that exist between;
Interactions(u0，uk) Representing the user u to be recommended0With social users ukThe number of social relationships that exist between;
Intimacy(u0，uj) Representing the user u to be recommended0With social users ujThe degree of intimacy therebetween.
In an optional embodiment, the calculation module is specifically configured to:
calculating the child intimacy degree of each social relationship between the user to be recommended and each social user in the social user set according to the type and the frequency of the social relationship between the user to be recommended and each social user in the social user set;
and calculating the intimacy between the user to be recommended and each social user in the social user set according to the child intimacy of the user to be recommended and each social user in the social user set under each social relationship.
In an optional embodiment, the recommendation module comprises:
the calculating sub-module is used for calculating the intimacy between each social user in the social user set and each merchant in the merchant set;
the determining sub-module is used for determining recommendation scores of all merchants in the merchant set according to the intimacy between the user to be recommended and all social users in the social user set and the intimacy between all social users in the social user set and all merchants in the merchant set;
and the recommending sub-module is used for recommending the interested commercial tenant to the user to be recommended from the commercial tenant set according to the recommending score of each commercial tenant in the commercial tenant set.
In an optional embodiment, the calculation sub-module is specifically configured to:
and calculating the intimacy between each social user in the social user set and each merchant in the merchant set according to the attribute information of the network behavior relationship between each social user in the social user set and each merchant in the merchant set.
In an optional embodiment, the calculation sub-module is specifically configured to:
and calculating the intimacy between each social user in the social user set and each merchant in the merchant set according to the type and/or the times of the network behavior relationship existing between each social user in the social user set and each merchant in the merchant set.
In an optional embodiment, the calculation sub-module is specifically configured to:
according to the formula
Figure BDA0001314690860000081
Calculating the intimacy between each social user in the social user set and each merchant in the merchant set;
ujrepresenting the jth social user in the set of social users, 1<＝j<N, N represents the number of social users in the set of social users;
sirepresenting the i-th merchant in the set of merchants, 1<＝i<M, representing the number of merchants in the set of merchants;
OrderShops(uj) Representing social users ujA set of merchants having a network behavioral relationship;
slrepresenting social users ujThe ith merchant in the set of merchants with network behavior relationship, 1<＝l<L denotes a social user ujThe number of merchants with network behavior relationships;
OrderTimes(ui，sl) Representing social users ujWith merchants slThe number of times there is a network behavior relationship between them;
OrderTimes(uj，si) Representing social users ujWith merchants siThe number of times there is a network behavior relationship between them;
Intimacy(uj，si) Representing social useHuu (household)jWith merchants siThe degree of intimacy therebetween.
In an optional embodiment, the determining sub-module is specifically configured to:
calculating the recommendation score of each merchant in the merchant set according to the following formula;
Figure BDA0001314690860000091
u0representing the user to be recommended;
Friends(u0) Representing the set of social users;
ujrepresenting the jth social user in the set of social users, 1<＝j<N, N represents the number of social users in the set of social users;
skrepresents the kth merchant in the set of merchants, 1<＝k<M, representing the number of merchants in the set of merchants;
Intimacy(u0，uj) Representing the user to be recommended and the social user uj(ii) the intimacy;
Intimacy(uj，sk) Representing social users ujWith merchants sk(ii) the intimacy;
RecScore(sk) Representing said merchant skThe recommendation score of (1).
In an optional embodiment, the determining sub-module is further configured to:
and determining the merchant set from merchants having network behavior relations with any social users in the social user set.
The embodiment of the application also provides an electronic device, which comprises a memory and a processor; the memory is used for storing one or more computer instructions, wherein the one or more computer instructions, when executed by the processor, can implement the steps of the recommendation method provided by the above method embodiments.
The embodiment of the present application further provides a computer-readable storage medium storing a computer program, where the computer program, when executed by a computer, implements the steps in the recommendation method provided by the above method embodiment.
In the embodiment of the application, a social user set corresponding to a user to be recommended is determined, the intimacy between the user to be recommended and each social user in the social user set is calculated, and then an interested business is recommended to the user to be recommended from a business set associated with the social user set based on the intimacy between the user to be recommended and each social user. According to the method and the device, the business is recommended based on the social contact user of the user to be recommended, the recommended business is from the business related to the social contact user, the reliability of the recommended business is high, the trust degree of the user to be recommended on the recommended business is increased, the participation degree of the user to be recommended on the recommended business is further improved, and the recommendation effect is fully played.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:
fig. 1 is a schematic flowchart of a recommendation method according to an embodiment of the present application;
FIG. 2 is a schematic flow chart illustrating a process for recommending interested merchants according to another embodiment of the present application;
FIG. 3 is a schematic diagram illustrating a relationship for determining a recommendation score for each merchant in a set of merchants based on two affinities, according to yet another embodiment of the present application;
fig. 4 is a schematic structural diagram of a recommendation device according to another embodiment of the present application;
fig. 5 is a schematic structural diagram of a recommendation device according to another embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the technical solutions of the present application will be described in detail and completely with reference to the following specific embodiments of the present application and the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
Fig. 1 is a schematic flowchart of a recommendation method according to an embodiment of the present application. As shown in fig. 1, the method includes:
101. and determining a social user set corresponding to the user to be recommended.
102. And calculating the intimacy between the user to be recommended and each social user in the social user set.
103. And recommending the interested commercial tenants to the user to be recommended from the commercial tenant set associated with the social user set according to the intimacy between the user to be recommended and each social user in the social user set.
During the use of mobile internet applications (apps), recommendations of relevant content to users are often required for various reasons. For convenience of description, in the embodiment of the present application, a user to whom content needs to be recommended is referred to as a user to be recommended. The content to be recommended to the user may vary according to the application scenario. The embodiment of the application is mainly suitable for application scenes with a plurality of merchants, and is convenient for users to choose commodities from proper merchants and recommend the merchants to the users. For example, the recommendation method provided by the embodiment of the application can be applied to shopping apps or takeaway apps provided by various large power suppliers.
Referring to step 101, in the recommendation process, a user to be recommended is determined first, and then a social user set corresponding to the user to be recommended is determined. The user to be recommended may be any user, and may be an old user, a new user or a potential user of the mobile internet App, for example. The social user set comprises one or more users having social relations with the users to be recommended. For convenience of description, a user having a social relationship with a user to be recommended may be referred to as a social user of the user to be recommended.
The generalized social relationship mainly refers to an associative relationship between users generated due to social activities between the users. The generalized social activities refer to behavioral activities in which users transmit information and communicate ideas in a certain manner (or tool) to achieve a certain purpose. In this embodiment, the social relationship between the users needs to be used, and in consideration of convenience in obtaining the social relationship between the users, the social relationship in this embodiment gives priority to the association relationship between the users generated by the network behavior of communicating information and communicating ideas in the internet manner to achieve a certain purpose between the users, so as to conveniently obtain the social relationship between the users through the internet, but is not limited to this definition. Generalized social relationships and social activities are still applicable to embodiments of the present application. For example, chat behaviors, praise behaviors, comment behaviors, share behaviors and the like among users belong to the category of social relationships.
Continuing with step 102, on the basis of the set of social users determined in step 101, the affinity between the user to be recommended and each social user in the set of social users is calculated. The affinity is used for representing the interactive heat between the user to be recommended and the social user. Generally, the higher the affinity between the user to be recommended and the social user, the more frequent the interaction between the user to be recommended and the social user is, and the higher the interaction popularity is. The higher the interaction heat between the user to be recommended and the social contact user is, the higher the trust between the user to be recommended and the social contact user is.
Continuing with step 103, based on the affinity calculated in step 102, an interested business may be recommended to the user to be recommended from the business set associated with the social user set according to the affinity between the user to be recommended and each social user in the social user set.
The business set associated with the social user set includes one or more businesses which are related to the social users in the social user set, for example, the businesses which the certain social user or certain social users purchased the goods, or the businesses which the certain social user or certain social users commented on, etc.
In the embodiment, the merchant recommendation is performed based on the social contact user of the user to be recommended, the recommended merchant is from a merchant associated with the social contact user, the reliability of the recommended merchant is high, the trust degree of the user to be recommended on the recommended merchant is increased, the participation degree of the user to be recommended on the recommended merchant is further improved, the recommendation effect is more fully exerted, for example, the purpose of increasing the number of users is achieved, or the purpose of improving the service quality is achieved.
In the above embodiment or the following embodiments, a social user set corresponding to a user to be recommended needs to be determined. In an optional embodiment, the social relationship of the user to be recommended may be obtained through social data of the user to be recommended. Based on the method, the social data of the user to be recommended can be captured, and the user having the social relationship with the user to be recommended is mined from the social data to form a social user set.
Generally, in the internet era, a user to be recommended often uses tools such as a social App, a social website and/or an application App to generate social activities. For example, common social apps have WeChat, QQ, etc.; common social websites include a human network, a microblog, a happy network and the like; common application apps include a take-out App, a shopping App, a live App, a game App, and the like. For example, a user may chat with friends, share information, view circles of friends or like through WeChat, QQ, and thus form social relationships with the friends. For another example, the user may discuss hot topics, make comments, post posts, and the like with other users through the micro blog, thereby forming a social relationship with other users. For another example, the user may consult a merchant for information, submit an order, comment on a product, store a favorite, and the like through a takeout App or a shopping App, and may inquire about suggestions and the like from other users who purchase the same product, thereby forming a social relationship with the merchant or other users. For another example, a user may interact with a main broadcast in a live broadcast room through a live broadcast App and interact with other users watching the same live broadcast, thereby forming a social relationship with the main broadcast or other users.
Based on the above, the social data of the user to be recommended on the social App, the social website and/or the application App can be captured; and mining users having social relations with the users to be recommended from the social data of the users to be recommended to form a social user set.
Further optionally, capturing social data of the user to be recommended on the social App, the social website and/or the application App, including: the method comprises the following steps of grabbing relevant data of at least one social behavior of a user to be recommended on a social App, a social website and/or an application App: the data comprises data related to chat behaviors, data related to sharing behaviors, data related to picking behaviors, data related to viewing behaviors, data related to commenting behaviors and data related to praise behaviors.
For example, the data related to the chat behavior may include identification information of the chat user, chat time, chat content, identification information of a chat tool (Wechat or QQ), and the like.
For example, the data related to the sharing behavior may include identification information of the sharing user, sharing time, sharing content, sharing manner, and the like.
For example, the data related to the pickup behavior may include identification information of a pickup user, pickup time, pickup content, identification information of a user sharing the pickup content, and the like.
For example, the data related to the viewing behavior may include identification information of the viewing user, viewing time, viewing content, viewing mode, information of a publisher viewing the content, and the like.
For example, the relevant data of the comment behavior may include identification information of the comment user, identification information of the object to be commented, comment time, comment content, and the like.
For example, the data related to the approval behavior may include identification information of an approved user, identification information of an approved object, approved content, approval time, approval approach (e.g., WeChat or QQ), and the like.
Further optionally, mining, from the social data, users who have a social relationship with the user to be recommended, and forming the social user set in one embodiment is: the social data are directly analyzed, social data containing identification information of the user to be recommended and identification information of other users are obtained, and the user identified by the identification information of the other users in the social data is taken as a social user of the user to be recommended and added to a social user set.
Further optionally, mining, from the social data, users who have a social relationship with the user to be recommended, and forming another embodiment of the social user set includes: constructing a social network graph of the user to be recommended by taking the users in the social data as vertexes and taking social relations among the users in the social data as edges; and acquiring users having social relations with the users to be recommended based on the social network diagram to form a social user set.
In a specific application example, suppose that a user to be recommended shares a red envelope through an application App, for example, a takeout App, and accordingly, other users can receive the red envelope shared by the user to be recommended, and social data is generated by the behavior that the user to be recommended shares the red envelope. These social data include, but are not limited to: identification information of the user to be recommended, shared red envelope content, identification information of the user who receives the red envelope, and the like. Based on the method, social data generated by sharing the red envelope on the application App by the user to be recommended can be captured, and a social network diagram (U, V, E) of the user to be recommended is constructed based on the social data. The user nodes comprise nodes corresponding to sharing users and nodes corresponding to receiving users; v is a receiving relation between the sharing user and the receiving user and is represented as an edge between a node corresponding to the sharing user and a node corresponding to the receiving user; and E is the value of the edge and is expressed as the receiving times. Furthermore, other nodes having edges with the node corresponding to the user to be recommended may be acquired from the social network diagram, and the user corresponding to the acquired other nodes is taken as the social user of the user to be recommended and added to the social user set.
It should be noted that, in the early stage of implementation of the recommendation method, the social network diagram may be an undirected diagram in consideration of sparseness of data, so as to solve the problem of sparseness of data and ensure that the recommendation method can be implemented normally. Of course, data is continuously accumulated along with the implementation of the recommendation method, and when the data is accumulated to a certain degree, the social network graph can be a directed graph so as to improve the accuracy of the recommendation result. Of course, the implementation form of the social network diagram may have no relation to the implementation time of the recommendation method. For example, throughout the implementation of the recommendation method, the social network diagram is always undirected; or, the social network graph is always a directed graph throughout the implementation of the recommendation method.
In the foregoing embodiment or the following embodiments, after determining the social user set corresponding to the user to be recommended, it is necessary to calculate the affinity between the user to be recommended and each social user in the social user set.
Alternatively, one way to calculate the intimacy is to: and calculating the intimacy between the user to be recommended and each social user in the social user set according to the attribute information of the social relationship between the user to be recommended and each social user in the social user set. It should be noted that, here, the attribute information of the social relationship existing between the user to be recommended and the social user has a one-to-one correspondence relationship with the affinity between the user to be recommended and the social user. In short, for a social user, the affinity between the user to be recommended and the social user needs to be calculated according to the attribute information of the social relationship existing between the user to be recommended and the social user.
The attribute information of the social relationship refers to several information used to describe the social relationship, and may be, for example, a type of the social relationship, a number of times of the social relationship, an occurrence time of the social relationship, and the like.
The type of social relationship may also vary, differentiated from different dimensions. For example, the type of social relationship may be of the social App type, or may be of the social website type, or may be of the take away App type, or may be of the game App type, and so forth, distinguished from the class of tools that generate the social relationship. Differentiated from the type of behavior of the user that generated the social relationship, the type of social relationship may be of the chat type, or may be of the shopping type, or may be of the like, or may be of the favorites type, or may be of the share type, or may be of the pick-up type, and so on.
For example, taking the attribute information of the social relationship including the type of the social relationship as an example, the affinity between the user to be recommended and each social user in the set of social users may be calculated according to the type of the social relationship existing between the user to be recommended and each social user in the set of social users.
In an example, a corresponding relationship between the type of the social relationship and the affinity may be preset, and when the affinity between the user to be recommended and any social user is calculated, the type of the social relationship existing between the user to be recommended and the social user may be matched in the corresponding relationship, and the affinity corresponding to the type of the social relationship in the matching may be used as the affinity between the user to be recommended and the social user.
In another example, priorities among different social relationship types may be preset, and then a corresponding relationship between the priorities and the affinities is set, so that when the affinities between the user to be recommended and any social user are calculated, the priorities of the types of social relationships existing between the user to be recommended and the social users may be determined, matching is performed in the corresponding relationship between the priorities and the affinities according to the determined priorities, and the affinities in the matching are used as the affinities between the user to be recommended and the social users.
For another example, taking the number of times that the attribute information of the social relationship includes the social relationship as an example, the affinity between the user to be recommended and each social user in the social user set may be calculated according to the number of times that the social relationship exists between the user to be recommended and each social user in the social user set. The more the number of times of the social relationship between the user to be recommended and the social user is, the higher the intimacy between the user to be recommended and the social user is; conversely, the lower the affinity between the user to be recommended and the social user.
In an example, when the affinity between the user to be recommended and each social user in the social user set is calculated according to the number of times of the social relationship between the user to be recommended and each social user in the social user set, the number of times of the social relationship between the user to be recommended and each social user in the social user set may be directly used as the affinity between the user to be recommended and each social user for any social user.
In another example, when the affinity between the user to be recommended and each social user in the social user set is calculated according to the number of times of the social relationship between the user to be recommended and each social user in the social user set, for any social user, the number of times of the social relationship between the user to be recommended and each social user in the social user set may be multiplied by a coefficient corresponding to the social user, and the multiplication result is used as the affinity between the user to be recommended and each social user.
In yet another example, when the affinity between the user to be recommended and each social user in the set of social users is calculated according to the number of times of the social relationship existing between the user to be recommended and each social user in the set of social users, the affinity between the user to be recommended and each social user in the set of social users may be calculated according to the following formula (1).
Figure BDA0001314690860000161
In the formula (1), u0Representing a user to be recommended; friends (u)0) Representing a social user set corresponding to a user to be recommended; u. ofjRepresenting the jth social user in the set of social users, 1<＝j<N denotes the number of social users in the set of social users; accordingly, ukRepresenting the kth social user in the set of social users, 1<＝k<＝N；Interactions(u0，uj) Representing a user u to be recommended0With social users ujThe number of social relationships that exist between; interactions (u)0，uk) Representing a user u to be recommended0With social users ukThe number of social relationships that exist between; intimacy (u)0，uj) Representing a user u to be recommended0With social users ujThe degree of intimacy therebetween.
The expression of formula (1) means: for any social user in the social user set, a ratio of the number of times of social relationships between the user to be recommended and the social user to the total number of times of social relationships between the user to be recommended and all the social users in the social user set may be calculated, where the ratio is the affinity between the user to be recommended and the social user.
Of course, in addition to the method of calculating the intimacy degree shown in the formula (1), other deformation formulas of the formula (1) may be used to calculate the intimacy degree.
For another example, taking the attribute information of the social relationship including the type and the number of times of the social relationship as an example, the affinity between the user to be recommended and each social user in the social user set may be calculated according to the type and the number of times of the social relationship existing between the user to be recommended and each social user in the social user set.
In one example, when the affinity between the user to be recommended and each social user in the social user set is calculated according to the type and the number of times of the social relationship existing between the user to be recommended and each social user in the social user set, for any social user, a weight value may be determined according to the type of the social relationship existing between the user to be recommended and each social user, the number of times of the social relationship existing between the user to be recommended and each social user is multiplied by the corresponding weight value, and the multiplication result is used as the affinity between the user to be recommended and each social user. Optionally, a corresponding relationship between the social relationship type and the weight value may be preset, and the weight value corresponding to the social relationship type is obtained based on the corresponding relationship. Alternatively, a conversion function between the social relationship type and the weight value may be preset, and the conversion is performed based on the conversion function to obtain the weight value corresponding to the social relationship type.
In another example, when the closeness between the user to be recommended and each social user in the social user set is calculated according to the type and the number of the social relationships between the user to be recommended and each social user in the social user set, for any social user, the child closeness between the user to be recommended and each social user in each social relationship can be calculated according to the type and the number of the social relationships between the user to be recommended and each social user; and calculating the intimacy between the user to be recommended and the social contact user according to the child intimacy of the user to be recommended and the social contact user under each social relationship.
When the above another example is implemented specifically, for any social user, the types of social relationships existing between the user to be recommended and the social user may be counted first, and then the number of times of the social relationships in each social relationship may be counted; for any social relationship, calculating the ratio of the number of times of the social relationship between the user to be recommended and the social user in the social relationship to the total number of times of the social relationship between the user to be recommended and all the social users in the social relationship, and taking the ratio as the child intimacy of the user to be recommended and the social user in the social relationship; then, the child intimacy of the user to be recommended and the social contact user under each social relationship is weighted, and accordingly the intimacy between the user to be recommended and the social contact user is obtained.
In the foregoing embodiment or the following embodiments, after obtaining the affinity between the user to be recommended and each social user in the social user set, according to the affinity between the user to be recommended and each social user in the social user set, an interested business is recommended to the user to be recommended from the business set associated with the social user set.
In an example, when an interested business is recommended, a social user with the affinity greater than a threshold value with the user to be recommended may be selected, or a social user with the affinity within a set range with the user to be recommended is selected, and a business which is associated with the selected social user in the business set is recommended to the user to be recommended as the interested business. For example, merchants that are present in association with the selected social user may be, but are not limited to, merchants that have been singled, liked, and/or commented by the selected social user.
In another example, when an interested business is recommended, the selection weight of the user to be recommended and a business related to each social user in the business set can be determined according to the intimacy between the user to be recommended and each social user in the social user set, and then the interested business is selected from the business set and recommended to the user to be recommended based on the selection weight.
In yet another example, the way to recommend an interested merchant, as shown in FIG. 2, includes the following steps:
1031. and calculating the intimacy between each social user in the social user set and each business in the business set.
1032. And determining the recommendation score of each merchant in the merchant set according to the intimacy between the user to be recommended and each social user in the social user set and the intimacy between each social user in the social user set and each merchant in the merchant set.
1033. And recommending the interested commercial tenants to the user to be recommended from the commercial tenant set according to the recommendation scores of the commercial tenants in the commercial tenant set.
Optionally, before step 1031, a merchant set may be determined from merchants that have a network behavior relationship with any social user in the social user set. For example, all merchants having a network behavior relationship with any social user in the set of social users may be added to the set of merchants. Or, part of merchants can be selected from merchants having network behavior relation with any social user in the social user set to join in the merchant set. The selection condition may depend on the application requirements and the application scenario, and is not limited herein.
For example, the merchants having a network behavior relationship with the social users may include, but are not limited to: merchants approved by the social users, merchants commented by the social users, merchants collected by the social users, merchants shared by the social users, merchants listed by the social users, merchants added with goods by the social users to the shopping cart and/or merchants browsed by the social users, and the like.
In the example shown in fig. 2, the affinity between each social user in the social user set and each merchant in the merchant set is calculated, and then the social users in the social user set are used as intermediate transitioners, recommendation scores of each merchant in the merchant set are obtained according to the affinity between the user to be recommended and each social user in the social user set and the affinity between each social user in the social user set and each merchant in the merchant set, and interested merchants are obtained from the merchant set according to the recommendation scores of each merchant.
Optionally, the affinity between each social user in the social user set and each merchant in the merchant set may be calculated according to attribute information of a network behavior relationship existing between each social user in the social user set and each merchant in the merchant set.
The attribute information of the network behavior relationship refers to several pieces of information used for describing the network behavior relationship, and may be, for example, a type of the network behavior relationship, a number of times of the network behavior relationship, an occurrence time of the network behavior relationship, and the like. Wherein, the type of the network behavior relationship can be determined by the type of the network behavior. For example, the type of network behavior relationship may be a type of placing an order, or may be a type of like, or may be a type of collection, or may be a type of sharing, or may be a type of comment, and so on.
For example, taking the attribute information of the network behavior relationship including the type of the network behavior relationship as an example, the affinity between each social user in the social user set and each merchant in the merchant set may be calculated according to the type of the network behavior relationship existing between each social user in the social user set and each merchant in the merchant set.
In an example, a corresponding relationship between the network behavior relationship type and the affinity may be preset, and when calculating the affinity between any social user and any merchant, the type of the network behavior relationship existing between the social user and the merchant may be matched in the corresponding relationship, and the affinity corresponding to the network behavior relationship type in the matching is used as the affinity between the social user and the merchant.
In another example, priorities among different types of network behavior relationships may be preset, and then a correspondence between the priorities and the affinities is set, so that when calculating the affinities between any social user and any business, the priorities of the types of network behavior relationships existing between the social user and the business may be determined, matching is performed in the correspondence between the priorities and the affinities according to the determined priorities, and the affinities in the matching are used as the affinities between the social user and the business.
For another example, taking the number of times that the attribute information of the network behavior relationship includes the network behavior relationship as an example, the affinity between each social user in the social user set and each merchant in the merchant set may be calculated according to the number of times that the network behavior relationship exists between each social user in the social user set and each merchant in the merchant set.
In an example, when the affinity between each social user in the social user set and each merchant in the merchant set is calculated according to the number of times of the network behavior relationship between each social user in the social user set and each merchant in the merchant set, for any social user and any merchant, the number of times of the social relationship between the social user and the merchant may be directly used as the affinity between the social user and the merchant.
In another example, when the affinity between each social user in the set of social users and each merchant in the set of merchants is calculated according to the number of times of the network behavior relationship existing between each social user in the set of social users and each merchant in the set of merchants, the affinity between each social user in the set of social users and each merchant in the set of merchants may be calculated according to the following formula (2).
Figure BDA0001314690860000201
In the formula (2), ujRepresenting the jth social user in the set of social users, 1<＝j<N denotes the number of social users in the set of social users; siRepresenting the ith merchant in the set of merchants, 1<＝i<M, which represents the number of merchants in the set of merchants; ordershop (u)j) Representing social users ujA set of merchants having a network behavioral relationship; slRepresenting social users ujThe ith merchant in the set of merchants with network behavior relationship, 1<＝l<L denotes a social user ujThe number of merchants with network behavior relationships; OrderTimes (u)i，sl) Representing social users ujWith merchants slThe number of times there is a network behavior relationship between them; OrderTimes(uj，si) Representing social users ujWith merchants siThe number of times there is a network behavior relationship between them; intimacy (u)j，si) Representing social users ujWith merchants siThe degree of intimacy therebetween.
The expression of formula (2) means: for any social user in the set of social users and any merchant in the set of merchants, a ratio of the number of times of the network behavior relationship existing between the social user and the merchant to the total number of times of the network behavior relationship existing between the social user and all merchants having the network behavior relationship with the social user may be calculated, where the ratio is the affinity between the social user and the merchant.
Of course, in addition to the method of calculating the intimacy degree shown in the formula (2), other deformation formulas of the formula (2) may be used to calculate the intimacy degree.
For another example, taking the attribute information of the network behavior relationship including the type and the number of times of the network behavior relationship as an example, the affinity between each social user in the social user set and each merchant in the merchant set may be calculated according to the type and the number of times of the network behavior relationship existing between each social user in the social user set and each merchant in the merchant set.
In an example, when the type and the number of times of the network behavior relationship existing between each social user in the social user set and each merchant in the merchant set are calculated, for any social user and any merchant, a weight value may be determined according to the type of the network behavior relationship existing between the social user and the merchant, and the number of times of the network behavior relationship existing between the social user and the merchant is multiplied by the corresponding weight value, and the multiplication result is used as the affinity between the social user and the merchant. Optionally, a correspondence between the network behavior relationship type and the weight value may be preset, and the weight value corresponding to the network behavior relationship type is obtained based on the correspondence. Alternatively, a conversion function between the network behavior relationship type and the weight value may be preset, and the conversion is performed based on the conversion function to obtain the weight value corresponding to the network behavior relationship type.
In another example, when calculating the affinity between each social user in the social user set and each merchant in the merchant set according to the type and the number of the network behavior relationships between each social user in the social user set and each merchant in the merchant set, for any social user and any merchant, calculating the child affinity of the social user and the merchant under each network behavior relationship according to the type and the number of the network behavior relationships between the social user and the merchant; and calculating the intimacy between the social user and the commercial tenant according to the child intimacy between the social user and the commercial tenant under each network behavior relationship.
In another embodiment, the type of the network behavior relationship existing between the social user and the merchant may be counted first, and then the number of times of the network behavior relationship under each network behavior relationship may be counted; for any network behavior relationship, calculating the ratio of the times of the network behavior relationship between the social user and the merchant under the network behavior relationship to the total times of the network behavior relationship between the social user and all merchants with the network behavior relationship between the social user and the merchant under the network behavior relationship, and taking the ratio as the child intimacy degree of the social user and the merchant under the network behavior relationship; then, the child affinity of the social user and the business under each network behavior relationship is weighted, so that the affinity between the social user and the business is obtained.
Optionally, in step 1032, a schematic diagram of determining the recommendation score of each merchant in the merchant set according to the affinity between the user to be recommended and each social user in the social user set and the affinity between each social user in the social user set and each merchant in the merchant set is shown in fig. 3. Optionally, one way to calculate the recommendation score of the merchant is:
the recommendation score for each merchant in the set of merchants is calculated according to the following formula (3).
Figure BDA0001314690860000221
In the formula (3), u0Representing a user to be recommended; friends (u)0) Representing a set of social users; u. ofjRepresenting the jth social user in the set of social users, 1<＝j<N denotes the number of social users in the set of social users; skRepresenting the kth merchant in the set of merchants, 1<＝k<M, which represents the number of merchants in the set of merchants; intimacy (u)0，uj) Representing users to be recommended and social users uj(ii) the intimacy; intimacy (u)j，sk) Representing social users ujWith merchants sk(ii) the intimacy; RecScore(s)k) Representing merchants skThe recommendation score of (1).
Based on the recommendation scores of the merchants calculated by the above formula (3), the merchants with recommendation scores greater than a set threshold value may be selected from the merchant set as interested merchants, or the merchants with recommendation scores within a specified score range may be selected from the merchant set as interested merchants, or the merchants may be ranked according to the recommendation scores of the merchants, and N merchants with the maximum recommendation scores are selected as interested merchants. After the interested merchant is determined, the interested merchant can be recommended to the user to be recommended.
In the embodiment of the application, the business is recommended based on the social contact user of the user to be recommended, the recommended business is from the business related to the social contact user, the reliability of the recommended business is high, the trust degree of the user to be recommended to the recommended business is increased, the participation degree of the user to be recommended to the recommended business is further improved, and the recommendation effect is more fully exerted. In addition, compared with the conventional recommendation method based on collaborative filtering, the embodiment of the application carries out merchant recommendation based on the social users of the users to be recommended, is beneficial to improving the diversity of recommendation results, overcomes the problem that the recommendation results are single or similar, and improves the participation of the users to a certain extent by means of richer recommendation results.
It should be noted that the execution subjects of the steps of the methods provided in the above embodiments may be the same device, or different devices may be used as the execution subjects of the methods. For example, the execution subjects of steps 101 to 103 may be device a; for another example, the execution subject of steps 101 and 102 may be device a, and the execution subject of step 103 may be device B; and so on.
Fig. 4 is a schematic structural diagram of a recommendation device according to another embodiment of the present application. As shown in fig. 4, the recommendation apparatus includes: a determination module 41, a calculation module 42 and a recommendation module 43.
The determining module 41 is configured to determine a social user set corresponding to the user to be recommended.
And the calculating module 42 is configured to calculate the affinity between the user to be recommended and each social user in the social user set determined by the determining module 41.
And the recommending module 43 is configured to recommend an interested business to the user to be recommended from the business set associated with the social user set according to the affinity between the user to be recommended and each social user in the social user set calculated by the calculating module 42.
In an alternative embodiment, as shown in fig. 5, one implementation structure of the determining module 41 includes: a grab sub-module 411 and a dig sub-module 412.
The crawling sub-module 411 is configured to crawl social data of the user to be recommended on the social App, the social website and/or the application App. Correspondingly, the mining submodule 412 is configured to mine, from the social data captured by the capture submodule 411, a user having a social relationship with the user to be recommended, so as to form a social user set.
Further, the grasping sub-module 411 is specifically configured to: the method comprises the following steps of grabbing relevant data of at least one social behavior of a user to be recommended on a social App, a social website and/or an application App:
the data comprises data related to chat behaviors, data related to sharing behaviors, data related to picking behaviors, data related to viewing behaviors, data related to commenting behaviors and data related to praise behaviors.
Further, the mining submodule 412 is specifically configured to: constructing a social network graph of the user to be recommended by taking the users in the social data as vertexes and taking social relations among the users in the social data as edges; and acquiring users having social relations with the users to be recommended based on the social network diagram to form a social user set.
In an alternative embodiment, the calculation module 42 is specifically configured to: and calculating the intimacy between the user to be recommended and each social user in the social user set according to the attribute information of the social relationship between the user to be recommended and each social user in the social user set.
Further, the calculation module 42 is specifically configured to: and calculating the intimacy between the user to be recommended and each social user in the social user set according to the type and/or the frequency of the social relationship between the user to be recommended and each social user in the social user set.
Further, when the affinity between the user to be recommended and each social user in the social user set is calculated according to the number of times of the social relationship existing between the user to be recommended and each social user in the social user set, the calculating module 42 is specifically configured to calculate the affinity between the user to be recommended and each social user in the social user set according to the formula (1). For formula (1), reference may be made to the description in the foregoing method embodiments, and details are not repeated here.
Further, when the affinity between the user to be recommended and each social user in the social user set is calculated according to the type and the number of times of the social relationship between the user to be recommended and each social user in the social user set, the calculating module 42 is specifically configured to: calculating the child intimacy of the user to be recommended and each social contact user in the social contact user set under each social contact relationship according to the type and the frequency of the social contact relationship existing between the user to be recommended and each social contact user in the social contact user set; and calculating the intimacy between the user to be recommended and each social user in the social user set according to the child intimacy between the user to be recommended and each social user in the social user set under each social relationship.
In an alternative embodiment, as shown in fig. 5, one implementation structure of the recommending module 43 includes: a calculation submodule 431, a determination submodule 432 and a recommendation submodule 433.
And the calculating submodule 431 is used for calculating the intimacy between each social user in the social user set and each business in the business set. The determining submodule 432 is configured to determine the recommendation score of each merchant in the merchant set according to the intimacy between the user to be recommended and each social user in the social user set and the intimacy between each social user in the social user set and each merchant in the merchant set, which is calculated by the calculating submodule 431. And the recommending submodule 433 is configured to recommend an interested merchant to the user to be recommended from the merchant set according to the recommendation score of each merchant in the merchant set determined by the determining submodule 432.
Further, the calculation submodule 431 is specifically configured to: and calculating the intimacy between each social user in the social user set and each merchant in the merchant set according to the attribute information of the network behavior relationship between each social user in the social user set and each merchant in the merchant set.
Further, the calculation submodule 431 is specifically configured to: and calculating the intimacy between each social user in the social user set and each merchant in the merchant set according to the type and/or the times of the network behavior relationship between each social user in the social user set and each merchant in the merchant set.
When the affinity between each social user in the social user set and each merchant in the merchant set is calculated according to the number of times of the network behavior relationship between each social user in the social user set and each merchant in the merchant set, the calculating sub-module 431 may specifically calculate the affinity between each social user in the social user set and each merchant in the merchant set according to the formula (2). For formula (2), reference may be made to the description in the foregoing method embodiments, and details are not repeated here.
Optionally, the determining submodule 432 is specifically configured to: and (4) calculating the recommendation score of each merchant in the merchant set according to the formula (3). For formula (3), reference may be made to the description in the foregoing method embodiments, and details are not repeated here.
Optionally, the determining submodule 431 is further configured to: before the calculating sub-module 431 calculates the affinity between each social user in the social user set and each business in the business set, the business set is determined from businesses which have network behavior relationship with any social user in the social user set.
For example, the determining sub-module 431 may add all of the merchants that have a network behavior relationship with any social user in the set of social users to the set of merchants. Alternatively, the determining sub-module 431 may select a part of the merchants from the merchants having network behavior relationship with any social user in the set of social users to join the set of merchants. The selection condition may depend on the application requirements and the application scenario, and is not limited herein.
The recommendation device provided in this embodiment may be configured to execute corresponding processes in the foregoing method embodiments, and details of a specific working principle of the recommendation device are not described again, and refer to the description in the foregoing method embodiments for details.
The recommending device provided by the embodiment recommends the merchants based on the social users of the users to be recommended, the recommended merchants are from the merchants associated with the social users, the reliability of the recommended merchants is high, the trust of the users to be recommended on the recommended merchants is increased, the participation of the users to be recommended on the recommended merchants is further improved, and the recommending effect is more fully exerted. In addition, compared with the existing recommendation method based on collaborative filtering, the recommendation device provided by the embodiment performs merchant recommendation based on the social users of the users to be recommended, so that the diversity of recommendation results is favorably improved, the problem that the recommendation results are single or similar is solved, and the participation of the users is improved to a certain extent by richer recommendation results.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The above description is only an example of the present application and is not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
The embodiment of the application discloses A1, a recommendation method, including:
determining a social user set corresponding to a user to be recommended;
calculating the intimacy between the user to be recommended and each social user in the social user set;
and recommending interest merchants to the user to be recommended from a merchant set associated with the social user set according to the intimacy between the user to be recommended and each social user in the social user set.
A2, the method A1, includes the following steps:
capturing social data of the user to be recommended on a social contact App, a social contact website and/or an application App;
and mining users having social relations with the users to be recommended from the social data to form the social user set.
A3, the method as A2, includes that the method includes that the social data of the user to be recommended on a social App, a social website and/or an application App are captured, and the method includes:
capturing relevant data of at least one social behavior of the user to be recommended on a social App, a social website and/or an application App:
the data comprises data related to chat behaviors, data related to sharing behaviors, data related to picking behaviors, data related to viewing behaviors, data related to commenting behaviors and data related to praise behaviors.
A4, the method as in A2, mining users who have social relations with the users to be recommended from the social data to form the social user set, including:
taking the users in the social data as vertexes and taking the social relations among the users in the social data as edges to construct a social network graph of the user to be recommended;
and acquiring the users with social relations with the users to be recommended based on the social network diagram to form the social user set.
A5, the method as in A1, wherein the calculating the affinity between the user to be recommended and each social user in the social user set comprises:
and calculating the intimacy between the user to be recommended and each social user in the social user set according to the attribute information of the social relationship existing between the user to be recommended and each social user in the social user set.
A6, the method as in A5, wherein the calculating the affinity between the user to be recommended and each social user in the set of social users according to the attribute information of the social relationship existing between the user to be recommended and each social user in the set of social users includes:
and calculating the intimacy between the user to be recommended and each social user in the social user set according to the type and/or the frequency of the social relationship between the user to be recommended and each social user in the social user set.
A7, the method as in A6, wherein the calculating the affinity between the user to be recommended and each social user in the set of social users according to the number of times of the social relationships between the user to be recommended and each social user in the set of social users includes:
according to the formula
Figure BDA0001314690860000291
Calculating the intimacy between the user to be recommended and each social user in the social user set;
u0representing the user to be recommended;
Friends(u0) Representing the set of social users;
ujrepresenting the jth social user in the set of social users, 1<＝j<N, N represents the number of social users in the set of social users;
ukrepresenting the kth social user of the set of social users, 1<＝k<＝N；
Interactions(u0，uj) Representing the user u to be recommended0With social users ujThe number of social relationships that exist between;
Interactions(u0，uk) Represents the user j to be recommended0With social users ukThe number of social relationships that exist between;
Intimacy(u0，uj) Representing the user u to be recommended0With social users ujThe degree of intimacy therebetween.
A8, the method as in A6, wherein the calculating the affinity between the user to be recommended and each social user in the set of social users according to the type and the number of social relationships comprises:
calculating the child intimacy degree of each social relationship between the user to be recommended and each social user in the social user set according to the type and the frequency of the social relationship between the user to be recommended and each social user in the social user set;
and calculating the intimacy between the user to be recommended and each social user in the social user set according to the child intimacy of the user to be recommended and each social user in the social user set under each social relationship.
In the method of a9, as in any one of a1-A8, recommending an interest business to the user to be recommended from a business set associated with the social user set according to an affinity between the user to be recommended and each social user in the social user set, the method includes:
calculating the intimacy between each social user in the social user set and each merchant in the merchant set;
determining recommendation scores of all merchants in the merchant set according to the intimacy between the user to be recommended and all social users in the social user set and the intimacy between all social users in the social user set and all merchants in the merchant set;
and recommending the interested commercial tenant to the user to be recommended from the commercial tenant set according to the recommendation score of each commercial tenant in the commercial tenant set.
A10, the method as in A9, wherein the calculating the affinity between each social user in the set of social users and each merchant in the set of merchants comprises:
and calculating the intimacy between each social user in the social user set and each merchant in the merchant set according to the attribute information of the network behavior relationship between each social user in the social user set and each merchant in the merchant set.
A11, the method as in A10, wherein the calculating the affinity between each social user in the set of social users and each business in the set of businesses according to the attribute information of the network behavior relationship existing between each social user in the set of social users and each business in the set of businesses includes:
and calculating the intimacy between each social user in the social user set and each merchant in the merchant set according to the type and/or the times of the network behavior relationship existing between each social user in the social user set and each merchant in the merchant set.
A12, the method as in A11, wherein the calculating the affinity between each social user in the set of social users and each business in the set of businesses according to the number of times of the network behavior relationship existing between each social user in the set of social users and each business in the set of businesses includes:
according to the formula
Figure BDA0001314690860000311
Calculating the intimacy between each social user in the social user set and each merchant in the merchant set;
ujrepresenting the jth social user in the set of social users, 1<＝j<N, N represents the number of social users in the set of social users;
sirepresenting the i-th merchant in the set of merchants, 1<＝i<M, representing the number of merchants in the set of merchants;
OrderShops(uj) Representing social users ujA set of merchants having a network behavioral relationship;
slrepresenting social users ujThe ith merchant in the set of merchants with network behavior relationship, 1<＝l<L denotes a social user ujThe number of merchants with network behavior relationships;
OrderTimes(ui，sl) Representing social users ujWith merchants slThe number of times there is a network behavior relationship between them;
OrderTimes(uj，si) Representing social users ujWith merchants siThe number of times there is a network behavior relationship between them;
Intimacy(uj，si) Representing social users ujWith merchants siThe degree of intimacy therebetween.
A13, in the method as in A9, determining the recommendation score of each merchant in the merchant set according to the affinity between the user to be recommended and each social user in the social user set and the affinity between each social user in the social user set and each merchant in the merchant set, including:
calculating the recommendation score of each merchant in the merchant set according to the following formula;
Figure BDA0001314690860000312
u0representing the user to be recommended;
Friends(u0) Representing the set of social users;
ujrepresenting the jth social user in the set of social users, 1<＝j<N, N represents the number of social users in the set of social users;
skrepresenting the kth quotient in the set of merchantsHouse 1<＝k<M, representing the number of merchants in the set of merchants;
Intimacy(u0，uj) Representing the user to be recommended and the social user uj(ii) the intimacy;
Intimacy(uj，sk) Representing social users ujWith merchants sk(ii) the intimacy;
RecScore(sk) Representing said merchant skThe recommendation score of (1).
A14, the method as in A9, before calculating the affinity between each social user in the set of social users and each merchant in the set of merchants, further comprising:
and determining the merchant set from merchants having network behavior relations with any social users in the social user set.
The embodiment of the application also discloses B15, a recommendation device, includes:
the determining module is used for determining a social user set corresponding to a user to be recommended;
the calculation module is used for calculating the intimacy between the user to be recommended and each social user in the social user set;
and the recommending module is used for recommending the interested commercial tenant to the user to be recommended from the commercial tenant set associated with the social user set according to the intimacy between the user to be recommended and each social user in the social user set.
B16, the apparatus of B15, wherein the determining module comprises: :
the crawling sub-module is used for crawling social data of the user to be recommended on the social App, the social website and/or the application App;
and the mining submodule is used for mining the users with social relations with the users to be recommended from the social data to form the social user set.
B17, the apparatus as defined in B16, wherein the grasping sub-module is specifically configured to:
capturing relevant data of at least one social behavior of the user to be recommended on a social App, a social website and/or an application App:
the data comprises data related to chat behaviors, data related to sharing behaviors, data related to picking behaviors, data related to viewing behaviors, data related to commenting behaviors and data related to praise behaviors.
B18, the apparatus as defined in B16, wherein the mining submodule is specifically configured to:
taking the users in the social data as vertexes and taking the social relations among the users in the social data as edges to construct a social network graph of the user to be recommended;
and acquiring the users with social relations with the users to be recommended based on the social network diagram to form the social user set.
B19, the apparatus of B15, wherein the computing module is specifically configured to:
and calculating the intimacy between the user to be recommended and each social user in the social user set according to the attribute information of the social relationship existing between the user to be recommended and each social user in the social user set.
B20, the apparatus of B19, wherein the computing module is specifically configured to:
and calculating the intimacy between the user to be recommended and each social user in the social user set according to the type and/or the frequency of the social relationship between the user to be recommended and each social user in the social user set.
B21, the apparatus of B20, wherein the computing module is specifically configured to:
according to the formula
Figure BDA0001314690860000331
Calculating the intimacy between the user to be recommended and each social user in the social user set;
u0representing the user to be recommended;
Friends(u0) Representing the set of social users;
ujrepresenting the social interactionThe jth social user in the user set, 1<＝j<N, N represents the number of social users in the set of social users;
ukrepresenting the kth social user of the set of social users, 1<＝k<＝N；
Interactions(u0，uj) Representing the user u to be recommended0With social users ujThe number of social relationships that exist between;
Interactions(u0，uk) Representing the user u to be recommended0With social users ukThe number of social relationships that exist between;
Intimacy(u0，uj) Representing the user u to be recommended0With social users ujThe degree of intimacy therebetween.
B22, the apparatus of B20, wherein the computing module is specifically configured to:
calculating the child intimacy degree of each social relationship between the user to be recommended and each social user in the social user set according to the type and the frequency of the social relationship between the user to be recommended and each social user in the social user set;
and calculating the intimacy between the user to be recommended and each social user in the social user set according to the child intimacy of the user to be recommended and each social user in the social user set under each social relationship.
B23, the device of any one of B15-B22, wherein the recommendation module comprises:
the calculating sub-module is used for calculating the intimacy between each social user in the social user set and each merchant in the merchant set;
the determining sub-module is used for determining recommendation scores of all merchants in the merchant set according to the intimacy between the user to be recommended and all social users in the social user set and the intimacy between all social users in the social user set and all merchants in the merchant set;
and the recommending sub-module is used for recommending the interested commercial tenant to the user to be recommended from the commercial tenant set according to the recommending score of each commercial tenant in the commercial tenant set.
B24, the apparatus as defined in B23, wherein the computing submodule is specifically configured to:
and calculating the intimacy between each social user in the social user set and each merchant in the merchant set according to the attribute information of the network behavior relationship between each social user in the social user set and each merchant in the merchant set.
B25, the apparatus as defined in B24, wherein the computing submodule is specifically configured to:
and calculating the intimacy between each social user in the social user set and each merchant in the merchant set according to the type and/or the times of the network behavior relationship existing between each social user in the social user set and each merchant in the merchant set.
B26, the apparatus as defined in B25, wherein the computing submodule is specifically configured to:
according to the formula
Figure BDA0001314690860000341
Calculating the intimacy between each social user in the social user set and each merchant in the merchant set;
ujrepresenting the jth social user in the set of social users, 1<＝j<N, N represents the number of social users in the set of social users;
sirepresenting the i-th merchant in the set of merchants, 1<＝i<M, representing the number of merchants in the set of merchants;
OrderShops(uj) Representing social users ujA set of merchants having a network behavioral relationship;
slrepresenting social users ujThe ith merchant in the set of merchants with network behavior relationship, 1<＝l<L denotes a social user ujThe number of merchants with network behavior relationships;
OrderTimes(ui，sl) Representing social users ujWith merchants slThe number of times there is a network behavior relationship between them;
OrderTimes(uj，si) Representing social users ujWith merchants siThe number of times there is a network behavior relationship between them;
Intimacy(uj，si) Representing social users ujWith merchants siThe degree of intimacy therebetween.
B27, the apparatus as defined in B23, wherein the determining submodule is specifically configured to:
calculating the recommendation score of each merchant in the merchant set according to the following formula;
Figure BDA0001314690860000351
u0representing the user to be recommended;
Friends(u0) Representing the set of social users;
ujrepresenting the jth social user in the set of social users, 1<＝j<N, N represents the number of social users in the set of social users;
skrepresents the kth merchant in the set of merchants, 1<＝k<M, representing the number of merchants in the set of merchants;
Intimacy(u0，uj) Representing the user to be recommended and the social user uj(ii) the intimacy;
Intimacy(uj，sk) Representing social users ujWith merchants sk(ii) the intimacy;
RecScore(sk) Representing said merchant skThe recommendation score of (1).
B28, the apparatus as described in B23, the determining sub-module further for:
and determining the merchant set from merchants having network behavior relations with any social users in the social user set.
The embodiment of the application also discloses C29, an electronic device, comprising a memory and a processor; the memory is configured to store one or more computer instructions, wherein the one or more computer instructions, when executed by the processor, are capable of performing the steps of any of the recommended methods provided by A1-A14, above.
The embodiment of the application also discloses D30 and a computer readable storage medium storing a computer program, wherein the computer program is used for realizing the steps in any recommendation method provided by A1-A14 when being executed by a computer.

Claims (26)

1. A recommendation method, comprising:
determining a social user set corresponding to a user to be recommended;
calculating the intimacy between the user to be recommended and each social user in the social user set;
recommending interested merchants to the user to be recommended from a merchant set associated with the social user set according to the intimacy between the user to be recommended and each social user in the social user set;
the recommending an interested business to the user to be recommended from a business set associated with the social user set according to the intimacy between the user to be recommended and each social user in the social user set comprises:
calculating the intimacy between each social user in the social user set and each merchant in the merchant set; determining recommendation scores of all merchants in the merchant set according to the intimacy between the user to be recommended and all social users in the social user set and the intimacy between all social users in the social user set and all merchants in the merchant set; recommending interesting merchants to the user to be recommended from the merchant set according to the recommendation scores of all merchants in the merchant set; wherein the set of merchants is determined from merchants having a network behavior relationship with social users in the set of social users;
determining recommendation scores of merchants in the merchant set according to the intimacy between the user to be recommended and each social user in the social user set and the intimacy between each social user in the social user set and each merchant in the merchant set, wherein the recommendation scores of merchants in the merchant set comprise:
calculating the recommendation score of each merchant in the merchant set according to the following formula;
Figure FDA0002357182380000011
u0representing the user to be recommended;
Friends(u0) Representing the set of social users;
ujrepresenting a jth social user in the set of social users, 1 ≦ j ≦ N, N representing a number of social users in the set of social users;
skrepresenting the kth merchant in the set of merchants, 1 < k < M, M representing the number of merchants in the set of merchants;
Intimacy(u0，uj) Representing the user to be recommended and the social user uj(ii) the intimacy;
Intimacy(uj，sk) Representing social users ujWith merchants sk(ii) the intimacy;
RecScore(sk) Representing said merchant skThe recommendation score of (1).
2. The method of claim 1, wherein determining the set of social users corresponding to the user to be recommended comprises:
capturing social data of the user to be recommended on a social contact App, a social contact website and/or an application App;
and mining users having social relations with the users to be recommended from the social data to form the social user set.
3. The method according to claim 2, wherein crawling social data of the user to be recommended on a social App, a social website and/or an application App comprises:
capturing relevant data of at least one social behavior of the user to be recommended on a social App, a social website and/or an application App:
the data comprises data related to chat behaviors, data related to sharing behaviors, data related to picking behaviors, data related to viewing behaviors, data related to commenting behaviors and data related to praise behaviors.
4. The method of claim 2, wherein mining users who have a social relationship with the user to be recommended from the social data to form the social user set comprises:
taking the users in the social data as vertexes and taking the social relations among the users in the social data as edges to construct a social network graph of the user to be recommended;
and acquiring the users with social relations with the users to be recommended based on the social network diagram to form the social user set.
5. The method of claim 1, wherein calculating the affinity between the user to be recommended and each social user in the set of social users comprises:
and calculating the intimacy between the user to be recommended and each social user in the social user set according to the attribute information of the social relationship existing between the user to be recommended and each social user in the social user set.
6. The method of claim 5, wherein calculating the affinity between the user to be recommended and each social user in the set of social users according to the attribute information of the social relationship existing between the user to be recommended and each social user in the set of social users comprises:
and calculating the intimacy between the user to be recommended and each social user in the social user set according to the type and/or the frequency of the social relationship between the user to be recommended and each social user in the social user set.
7. The method of claim 6, wherein calculating the affinity between the user to be recommended and each social user in the set of social users according to the number of times of the social relationship existing between the user to be recommended and each social user in the set of social users comprises:
according to the formula
Figure FDA0002357182380000031
Calculating the intimacy between the user to be recommended and each social user in the social user set;
u0representing the user to be recommended;
Friends(u0) Representing the set of social users;
ujrepresenting a jth social user in the set of social users, 1 ≦ j ≦ N, N representing a number of social users in the set of social users;
ukrepresenting a kth social user of the set of social users, 1 ≦ k ≦ N;
Interactions(u0，uj) Representing the user u to be recommended0With social users ujThe number of social relationships that exist between;
Interactions(u0，uk) Representing the user u to be recommended0With social users ukThe number of social relationships that exist between;
Intimacy(u0，uj) Representing the user u to be recommended0With social users ujThe degree of intimacy therebetween.
8. The method of claim 6, wherein calculating the affinity between the user to be recommended and each social user in the set of social users according to the type and the number of social relationships comprises:
calculating the child intimacy degree of each social relationship between the user to be recommended and each social user in the social user set according to the type and the frequency of the social relationship between the user to be recommended and each social user in the social user set;
and calculating the intimacy between the user to be recommended and each social user in the social user set according to the child intimacy of the user to be recommended and each social user in the social user set under each social relationship.
9. The method of claim 1, wherein calculating the affinity between each social user in the set of social users and each merchant in the set of merchants comprises:
and calculating the intimacy between each social user in the social user set and each merchant in the merchant set according to the attribute information of the network behavior relationship between each social user in the social user set and each merchant in the merchant set.
10. The method of claim 9, wherein calculating the affinity between each social user in the set of social users and each merchant in the set of merchants according to the attribute information of the network behavior relationship existing between each social user in the set of social users and each merchant in the set of merchants comprises:
and calculating the intimacy between each social user in the social user set and each merchant in the merchant set according to the type and/or the times of the network behavior relationship existing between each social user in the social user set and each merchant in the merchant set.
11. The method of claim 10, wherein calculating the affinity between each social user in the set of social users and each merchant in the set of merchants according to the number of times of the network behavior relationship existing between each social user in the set of social users and each merchant in the set of merchants comprises:
according to the formula
Figure FDA0002357182380000041
Calculating the intimacy between each social user in the social user set and each merchant in the merchant set;
ujrepresenting a jth social user in the set of social users, 1 ≦ j ≦ N, N representing a number of social users in the set of social users;
sirepresenting the ith merchant in the set of merchants, 1 < i < M, M representing the number of merchants in the set of merchants;
OrderShops(uj) Representing social users ujA set of merchants having a network behavioral relationship;
slrepresenting social users ujThe ith merchant in the set of merchants with network behavior relationship, 1 ≦ L, where L represents the relationship with the social user ujThe number of merchants with network behavior relationships;
OrderTimes(ui，sl) Representing social users ujWith merchants slThe number of times there is a network behavior relationship between them;
OrderTimes(uj，si) Representing social users ujWith merchants siThe number of times there is a network behavior relationship between them;
Intimacy(uj，si) Representing social users ujWith merchants siThe degree of intimacy therebetween.
12. The method of claim 1, wherein prior to calculating the affinity between each social user in the set of social users and each merchant in the set of merchants, further comprising:
and determining the merchant set from merchants having network behavior relations with any social users in the social user set.
13. A recommendation device, comprising:
the determining module is used for determining a social user set corresponding to a user to be recommended;
the calculation module is used for calculating the intimacy between the user to be recommended and each social user in the social user set;
the recommending module is used for recommending an interested business to the user to be recommended from a business set associated with the social user set according to the intimacy between the user to be recommended and each social user in the social user set;
the recommendation module comprises:
the calculating sub-module is used for calculating the intimacy between each social user in the social user set and each merchant in the merchant set;
the determining sub-module is used for determining recommendation scores of all merchants in the merchant set according to the intimacy between the user to be recommended and all social users in the social user set and the intimacy between all social users in the social user set and all merchants in the merchant set;
the recommending sub-module is used for recommending interesting merchants to the user to be recommended from the merchant set according to the recommendation scores of all merchants in the merchant set; wherein the set of merchants is determined from merchants having a network behavior relationship with social users in the set of social users;
the determination submodule is specifically configured to:
calculating the recommendation score of each merchant in the merchant set according to the following formula;
Figure FDA0002357182380000061
u0representing the user to be recommended;
Friends(u0) Representing the set of social users;
ujrepresenting a jth social user in the set of social users, 1 ≦ j ≦ N, N representing a number of social users in the set of social users；
skRepresenting the kth merchant in the set of merchants, 1 < k < M, M representing the number of merchants in the set of merchants;
Intimacy(u0，uj) Representing the user to be recommended and the social user uj(ii) the intimacy;
Intimacy(uj，sk) Representing social users ujWith merchants sk(ii) the intimacy;
RecScore(sk) Representing said merchant skThe recommendation score of (1).
14. The apparatus of claim 13, wherein the determining module comprises:
the crawling sub-module is used for crawling social data of the user to be recommended on the social App, the social website and/or the application App;
and the mining submodule is used for mining the users with social relations with the users to be recommended from the social data to form the social user set.
15. The apparatus of claim 14, wherein the grasping submodule is specifically configured to:
capturing relevant data of at least one social behavior of the user to be recommended on a social App, a social website and/or an application App:
the data comprises data related to chat behaviors, data related to sharing behaviors, data related to picking behaviors, data related to viewing behaviors, data related to commenting behaviors and data related to praise behaviors.
16. The apparatus of claim 14, wherein the mining submodule is specifically configured to:
taking the users in the social data as vertexes and taking the social relations among the users in the social data as edges to construct a social network graph of the user to be recommended;
and acquiring the users with social relations with the users to be recommended based on the social network diagram to form the social user set.
17. The apparatus of claim 13, wherein the computing module is specifically configured to:
and calculating the intimacy between the user to be recommended and each social user in the social user set according to the attribute information of the social relationship existing between the user to be recommended and each social user in the social user set.
18. The apparatus of claim 17, wherein the computing module is specifically configured to:
and calculating the intimacy between the user to be recommended and each social user in the social user set according to the type and/or the frequency of the social relationship between the user to be recommended and each social user in the social user set.
19. The apparatus of claim 18, wherein the computing module is specifically configured to:
according to the formula
Figure FDA0002357182380000071
Calculating the intimacy between the user to be recommended and each social user in the social user set;
u0representing the user to be recommended;
Friends(u0) Representing the set of social users;
ujrepresenting a jth social user in the set of social users, 1 ≦ j ≦ N, N representing a number of social users in the set of social users;
ukrepresenting a kth social user of the set of social users, 1 ≦ k ≦ N;
Interactions(u0，uj) Representing the user u to be recommended0With social users ujNumber of social relationships that exist between；
Interactions(u0，uk) Representing the user u to be recommended0With social users ukThe number of social relationships that exist between;
Intimacy(u0，uj) Representing the user u to be recommended0With social users ujThe degree of intimacy therebetween.
20. The apparatus of claim 18, wherein the computing module is specifically configured to:
calculating the child intimacy degree of each social relationship between the user to be recommended and each social user in the social user set according to the type and the frequency of the social relationship between the user to be recommended and each social user in the social user set;
and calculating the intimacy between the user to be recommended and each social user in the social user set according to the child intimacy of the user to be recommended and each social user in the social user set under each social relationship.
21. The apparatus of claim 13, wherein the computation submodule is specifically configured to:
and calculating the intimacy between each social user in the social user set and each merchant in the merchant set according to the attribute information of the network behavior relationship between each social user in the social user set and each merchant in the merchant set.
22. The apparatus of claim 21, wherein the computation submodule is specifically configured to:
and calculating the intimacy between each social user in the social user set and each merchant in the merchant set according to the type and/or the times of the network behavior relationship existing between each social user in the social user set and each merchant in the merchant set.
23. The apparatus of claim 22, wherein the computation submodule is specifically configured to:
according to the formula
Figure FDA0002357182380000081
Calculating the intimacy between each social user in the social user set and each merchant in the merchant set;
ujrepresenting a jth social user in the set of social users, 1 ≦ j ≦ N, N representing a number of social users in the set of social users;
sirepresenting the ith merchant in the set of merchants, 1 < i < M, M representing the number of merchants in the set of merchants;
OrderShops(uj) Representing social users ujA set of merchants having a network behavioral relationship;
slrepresenting social users ujThe ith merchant in the set of merchants with network behavior relationship, 1 ≦ L, where L represents the relationship with the social user ujThe number of merchants with network behavior relationships;
OrderTimes(ui，sl) Representing social users ujWith merchants slThe number of times there is a network behavior relationship between them;
OrderTimes(uj，si) Representing social users ujWith merchants siThe number of times there is a network behavior relationship between them;
Intimacy(uj，si) Representing social users ujWith merchants siThe degree of intimacy therebetween.
24. The apparatus of claim 13, wherein the determination sub-module is further configured to:
and determining the merchant set from merchants having network behavior relations with any social users in the social user set.
25. An electronic device comprising a memory and a processor; the memory is configured to store one or more computer instructions, wherein the one or more computer instructions, when executed by the processor, are capable of implementing the method as claimed in any one of claims 1-12.
26. A computer-readable storage medium storing a computer program, wherein the computer program is adapted to implement the recommendation method of any one of claims 1-12 when executed by a computer.
CN201710419957.3A 2017-06-06 2017-06-06 Recommendation method and device Active CN107436914B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710419957.3A CN107436914B (en) 2017-06-06 2017-06-06 Recommendation method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710419957.3A CN107436914B (en) 2017-06-06 2017-06-06 Recommendation method and device

Publications (2)

Publication Number Publication Date
CN107436914A CN107436914A (en) 2017-12-05
CN107436914B true CN107436914B (en) 2020-06-23

Family

ID=60459040

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710419957.3A Active CN107436914B (en) 2017-06-06 2017-06-06 Recommendation method and device

Country Status (1)

Country Link
CN (1) CN107436914B (en)

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109087183A (en) * 2018-09-28 2018-12-25 深圳壹账通智能科技有限公司 Products Show method, apparatus, computer readable storage medium and computer equipment
CN109684507A (en) * 2018-11-23 2019-04-26 北京达佳互联信息技术有限公司 Video recommendation method, video recommendations device and computer readable storage medium
CN109769128B (en) * 2018-12-25 2021-07-13 北京达佳互联信息技术有限公司 Video recommendation method, video recommendation device and computer-readable storage medium
CN110060053B (en) * 2019-01-30 2023-08-01 创新先进技术有限公司 Identification method, equipment and computer readable medium
CN111191143B (en) * 2019-07-17 2023-10-24 腾讯科技（深圳）有限公司 Application recommendation method and device
CN112861015A (en) * 2019-11-27 2021-05-28 北京达佳互联信息技术有限公司 Account associated information acquisition method and device in application program and electronic equipment
CN113297504A (en) * 2020-02-05 2021-08-24 北京达佳互联信息技术有限公司 Social account recommendation method and device, electronic equipment and storage medium

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102880691B (en) * 2012-09-19 2015-08-19 北京航空航天大学深圳研究院 A kind of mixing commending system based on user's cohesion and method
CN104021230B (en) * 2014-06-27 2017-02-15 重庆邮电大学 Collaborative filtering method based on community discovery
JP6457358B2 (en) * 2015-09-04 2019-01-23 株式会社東芝 Item recommendation system, item recommendation method and program
CN106528643B (en) * 2016-10-13 2020-10-16 上海师范大学 Multi-dimensional comprehensive recommendation method based on social network

Also Published As

Publication number Publication date
CN107436914A (en) 2017-12-05

Similar Documents

Publication Publication Date Title
CN107436914B (en) Recommendation method and device
JP7105700B2 (en) Time-division recommendation method and apparatus for service target
US11042946B2 (en) Identity mapping between commerce customers and social media users
US11176586B2 (en) Data analysis method and system thereof
US20140279233A1 (en) System and method for an affinity capture, user feedback and affinity analysis
US20130297382A1 (en) Network transaction platform and processing method thereof
CN110335123B (en) Commodity recommendation method, system, computer readable medium and device based on social e-commerce platform
CN110084658B (en) Method and device for matching articles
TWI705411B (en) Method and device for identifying users with social business characteristics
CN109493113A (en) A kind of providing method and device of favor information
CN108550046A (en) A kind of resource and market recommendation method, apparatus and electronic equipment
JP6199958B2 (en) User recommended methods and equipment
CN110647683A (en) Information recommendation method and device
US20110004508A1 (en) Method and system of generating guidance information
US20150302088A1 (en) Method and System for Providing Personalized Content
US8725795B1 (en) Content segment optimization techniques
US10474670B1 (en) Category predictions with browse node probabilities
US20120144413A1 (en) Ranking content using user feedback
CN110427545A (en) A kind of information-pushing method and system
CN107679236A (en) A kind of hot content pond maintaining method and device
CN113763089A (en) Article recommendation method and device and computer-readable storage medium
CN112184300A (en) Person-to-person matching method, medium, system and equipment
CN110020118B (en) Method and device for calculating similarity between users
RU2648951C1 (en) System and method of selecting and displaying recommended content to a user
CN114254204A (en) Data processing method, device, medium and electronic equipment

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
CB02 Change of applicant information

Address after: Building N3, Jinyu Zhizao workshop, No. 27, Jiancai Chengzhong Road, Haidian District, Beijing 100096

Applicant after: Beijing Xingxuan Technology Co.,Ltd.

Address before: 100085 Beijing, Haidian District on the road to the information on the ground floor of the 1 to the 3 floor of the 2 floor, room 11, 202

Applicant before: Beijing Xiaodu Information Technology Co.,Ltd.

CB02 Change of applicant information
GR01 Patent grant
GR01 Patent grant