WO2014000431A1 - 信息推荐方法和装置 - Google Patents

信息推荐方法和装置 Download PDF

Info

Publication number
WO2014000431A1
WO2014000431A1 PCT/CN2013/070158 CN2013070158W WO2014000431A1 WO 2014000431 A1 WO2014000431 A1 WO 2014000431A1 CN 2013070158 W CN2013070158 W CN 2013070158W WO 2014000431 A1 WO2014000431 A1 WO 2014000431A1
Authority
WO
WIPO (PCT)
Prior art keywords
user
information
social
circle
network platform
Prior art date
Application number
PCT/CN2013/070158
Other languages
English (en)
French (fr)
Inventor
丘志宏
齐泉
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Publication of WO2014000431A1 publication Critical patent/WO2014000431A1/zh
Priority to US14/333,784 priority Critical patent/US20140330653A1/en
Priority to US17/109,629 priority patent/US20210133817A1/en

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION 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
    • G06Q30/00Commerce
    • G06Q30/02Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241Advertisements
    • G06Q30/0251Targeted advertisements
    • G06Q30/0269Targeted advertisements based on user profile or attribute
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/953Querying, e.g. by the use of web search engines
    • G06F16/9535Search customisation based on user profiles and personalisation
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION 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
    • G06Q50/00Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/01Social networking
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04SSYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S50/00Market activities related to the operation of systems integrating technologies related to power network operation or related to communication or information technologies
    • Y04S50/14Marketing, i.e. market research and analysis, surveying, promotions, advertising, buyer profiling, customer management or rewards

Definitions

  • the present invention relates to the field of Internet technologies, and in particular, to an information recommendation method and apparatus. Background technique
  • Personalized recommendation technology is an important technology in the Internet field, especially in e-commerce. It can recommend users' information and products to users according to their interests and purchasing power.
  • the personalized recommendation engine is an intelligent platform based on massive data mining to help e-commerce websites and Internet information provision websites provide their users with fully personalized decision support and information services.
  • Content-based recommendation refers to discovering the relevance of an item or information based on the metadata of the recommended item or content, and recommending the item or information related to its historical interest to the user. For example, the e-commerce website finds through the user purchase record that User A has historically purchased history books, and User A has not purchased the current best-selling history book "Item 3", thus speculating that User A is the "Item 3" potential. The user then recommends "item 3" to user A.
  • Collaborative recommendation refers to the discovery of the user's relevance through the user's historical behavior record, based on the interests of other users related to the user.
  • the e-commerce website finds through the user purchase record that user A and user C always purchase the same item in history, so it is inferred that user A and user C have similar interests and interests; through the user purchase record, user A is also found to have purchased the item. 1", and user C has not purchased yet, so it is speculated that user C is a potential user of "item 1", and "item 1" is recommended to user C.
  • the embodiment of the invention provides a method and device for recommending information, which can utilize the open interface of the social networking website and user data for information recommendation, improve the accuracy of information recommendation, and provide great convenience for the user.
  • a first aspect of the present invention provides a method for recommending information, including: acquiring, by using an open interface of a first network platform, relationship data information of a user associated with each user in the second network platform, where the relationship data is The information includes user interaction information for interaction between users and user information indicating user behavior of the user;
  • information recommendation is separately performed in each of the social networks by using a preset recommendation strategy.
  • an information recommendation apparatus including: an obtaining module, configured to acquire, by using an open interface of the first network platform, relationship data information of a user associated with each user in the second network platform,
  • the relationship data information includes user interaction information exchanged between users and user behavior information indicating a user's own behavior;
  • a dividing module configured to divide each friend circle divided according to a preset dividing policy according to the relationship data information of the user, and divide a friend circle into a plurality of different social groups;
  • the recommendation module is configured to perform information recommendation in each of the social circles according to the obtained behavior record of each user in the second network platform by using a preset recommendation policy.
  • the technical effect of the embodiment of the present invention is: acquiring, by using an open interface of the first network platform, relationship data information of a user associated with each user in the second network platform, and dividing each friend circle into multiple different ones according to the relationship data information.
  • the social circle according to the behavior record of the user in the second network platform, performs information recommendation in the divided social circle; the embodiment can utilize the open interface of the social website and user data to perform information recommendation, thereby improving the accuracy of information recommendation. Degree, providing users with great convenience.
  • Embodiment 1 is a flowchart of Embodiment 1 of an information recommendation method according to the present invention.
  • FIG. 2 is a schematic diagram showing the relationship between a friend circle and a social circle in the first embodiment of the information recommendation method of the present invention
  • Embodiment 2 is a flowchart of Embodiment 2 of an information recommendation method according to the present invention.
  • FIG. 4 is a schematic diagram of a social circle-based collaborative recommendation process in Embodiment 2 of the information recommendation method of the present invention.
  • FIG. 5 is a schematic diagram of a system architecture in Embodiment 2 of the information recommendation method according to the present invention
  • FIG. 6 is a flowchart of Embodiment 3 of the information recommendation method according to the present invention
  • FIG. 7 is a schematic diagram of a content recommendation process based on a social circle in Embodiment 3 of the information recommendation method of the present invention.
  • Embodiment 8 is a schematic structural diagram of Embodiment 1 of an information recommendation apparatus according to the present invention.
  • FIG. 9 is a schematic structural diagram of Embodiment 2 of the information recommendation apparatus according to the present invention. detailed description
  • the e-commerce website uses the historical data generated by itself to make recommendations
  • the historical data and the potential user information and the information of the items to be recommended are highly consistent, and the potential users are the items purchased by the website.
  • the user, the item to be recommended is the same or very similar to the item purchased by the user in the historical data. Therefore, it is necessary to combine the historical data generated by the operator website or the social networking website to conduct information recommendation of the e-commerce website, so as to overcome the defect that the recommended item or information is not valuable to the user, and the traditional e-commerce based website
  • the method of recommending the historical data generated by itself cannot be directly applied to the recommendation based on social data generated by the operator's website or social networking website.
  • the present invention aims to solve the above technical problems, and proposes an information recommendation method, which utilizes resources opened by an operator or a social networking service provider to accurately recommend an item or information.
  • FIG. 1 is a flowchart of Embodiment 1 of the information recommendation method of the present invention. As shown in FIG. 1 , this embodiment provides an information recommendation method, which may specifically include the following steps:
  • Step 101 Acquire, by the open interface of the first network platform, relationship data information of a user associated with each user in the second network platform.
  • the first network platform in this embodiment may be specifically an operator website or a social networking website, such as facebook, twitter, Sina Weibo, etc.
  • the second network platform may be specifically an e-commerce website, such as Taobao, Jingdong Mall, Dangdang, etc.
  • the first network platform opens an API interface, and the second network platform can obtain the user relationship data information through the open interface of the first network platform.
  • the user here is a user associated with a user in the second network platform, where the association refers to a user having the same identity information on two network platforms, for example, the same person uses the same or different account in the first network.
  • the platform and the second network platform are registered to become users of the two network platforms.
  • the first network platform and the second network platform are two independent platforms, each having its own user, since the first network platform has an open interface, the second network platform can find a user of two platforms through the open interface.
  • the associated method that is, the second network platform can identify the user's identity from the registration information through the registration information of the user opened by the first network platform, such as an email address, and then obtain the registration information of the user of the second network platform itself. Don't identify the user in the second network platform. If the two users in the two network platforms have the same identity, the two users are the associated users.
  • the relationship data information herein includes user interaction information and user behavior information indicating the user's own behavior, and the user interaction information may be an email and a short message between the user and the friend in the first network platform.
  • Behavior or mutual browsing, forwarding, commenting, etc. between friends in blogs and microblogs; and this These behavior-related text data can be used to further divide the user's circle of friends.
  • the user behavior information may be a blog, a microblog, and the like published by the user, and may be used to determine a user's personal preference attribute.
  • Step 102 Divide each friend ⁇ that is divided according to the preset division policy according to the relationship data information of the user, and divide a friend circle into a plurality of different social circles.
  • each of the circle of friends may be separately divided according to the relationship data, and a friend is divided into a plurality of different social circles.
  • the buddy circle is divided according to the preset division policy.
  • the division strategy can be a user-centric division strategy or a division strategy according to the aggregation degree of the network.
  • the preset division strategy can be adopted.
  • the users in the network are divided into multiple friends, and different friends may usually include one or several identical users, that is, there are overlapping situations between different friends.
  • each of the friends is further divided, that is, the user can determine the user through the interaction between the friends and the communication discussion topic that the user publishes or participates in.
  • FIG. 2 is a schematic diagram of a relationship between a friend circle and a social network in the first embodiment of the information recommendation method according to the present invention.
  • a user circle of a user is divided into four social circles, namely, a technical assistant, a colleague, and a family circle. And outdoor activities.
  • Step 103 Perform information recommendation in each social circle by using a preset recommendation policy according to the obtained behavior record of each user in the second network platform.
  • this step uses the preset recommendation strategy to separately recommend information in each social circle.
  • the recommendation strategy here may be specifically a collaborative recommendation strategy, or a content recommendation strategy, or a collaborative recommendation strategy and content recommendation.
  • the information recommendation is performed by using a preset recommendation policy according to the obtained behavior record of each user in the social circle in the second network platform.
  • the behavior record of the user here in the second network platform may include a purchase record of the item of the user in the second network platform, a browsing record of the information, and the like. Since the interests of each user in a social circle are similar, the topics they care about or care about are similar.
  • the items or information with higher popularity are recommended, and other users in the circle usually refer to the recommended items.
  • the information is of interest, thereby improving the accuracy of the recommendation, and the user can obtain the item or information of interest without blind search.
  • the user has provided convenience.
  • the embodiment provides an information recommendation method, which acquires relationship data information of users associated with each user in the second network platform by using an open interface of the first network platform, and respectively divides each friend ⁇ according to the relationship data information. According to the behavior record of the user in the second network platform, the information recommendation is performed in the divided social network; the embodiment can utilize the open interface of the social website and the user data to perform information recommendation, and improve the information recommendation. The accuracy of the user provides great convenience.
  • FIG. 3 is a flowchart of Embodiment 2 of the information recommendation method of the present invention. As shown in FIG. 3, this embodiment provides an information recommendation method, which may specifically include the following steps:
  • Step 301 Acquire, by the open interface of the first network platform, relationship data information of a user associated with each user in the second network platform.
  • the first network platform opens the API interface outward, and the second network platform can obtain the user relationship data information through the open interface of the first network platform.
  • the user here is the user associated with the user in the second network platform, where the association refers to the user with the same identity information on both network platforms.
  • the first network platform and the second network platform are two independent platforms, each having its own user, since the first network platform has an open interface, the second network platform can find a user of two platforms through the open interface.
  • the associated method that is, the second network platform can identify the identity of the user from the registration information through the registration information of the user opened by the first network platform, and identify the second network platform by using the registration information of the user of the second network platform itself. The identity of the user. If the two users in the two network platforms have the same identity, the two users are the associated users.
  • Step 302 Acquire each user circle in the second network platform according to the relationship data information of each user.
  • the embodiment uses the information obtained from the first network platform.
  • the household's relationship data information is analyzed to accurately identify potential users.
  • users are a huge network of relationships.
  • the social data user of each user in the second network platform is obtained according to the relationship data information of each user acquired in the foregoing steps, where the social user of the user is a user who has social relationship with each user, and the social relationship is specific. Refers to the exchange of questions, mutual comments, and forwarding of Weibo between users through the first network platform.
  • the user and the social user of the user are divided into the circle of friends corresponding to the user, that is, a user who has a social relationship with the user is grouped into a circle of friends with the user as the center.
  • a friend circle may include multiple layers of friend relationships, for example, a two-layer friend relationship is: Assume that user A is the center, user B is the friend of user A, and user C is the friend of user B, and user C is also added. Go to the friend ⁇ corresponding to user A.
  • the embodiment may also form a circle of friends according to the degree of aggregation of the social network, that is, a node that is closely connected to each other in the social network may form a subnet, and the subnet is a good circle.
  • the social network here may be a network formed according to the relationship between users, each node in the network represents each user, and two nodes in the network are connected to each other to indicate that there is interaction between the two users, such as mutual Browse, forward, and other behaviors.
  • Step 303 Divide the circle of friends corresponding to each user according to the relationship data information of the user, and divide a circle of friends into a plurality of different social circles.
  • This step is to further filter the user's friends to more accurately identify the potential customers after the user circle is divided.
  • each of the friends is further divided, that is, the user can determine the user through the interaction between the friends and the communication discussion topic that the user publishes or participates in. Relationships with friends, such as classmates, colleagues, family members, academic circles on a topic, or discussion circles, can divide a circle of friends into multiple different social circles.
  • a user's circle of friends is divided into four social networks, namely technology circle, colleague circle, family circle and outdoor activities, and the users in each divided social circle are Can be considered as a potential customer of a certain category or a certain commodity or information.
  • Step 304 Obtain a behavior record of each user in the social circle in the second network platform.
  • the present embodiment performs information recommendation based on each social network.
  • the content recommendation policy and/or the collaborative recommendation policy may be used for recommendation.
  • the collaborative recommendation policy is taken as an example for description. This step is described by taking the information recommendation process in a social circle as an example. First, the behavior record of each user in the social circle in the second network platform is obtained, where the behavior record includes the purchase record of the item and the browsing record of the information. .
  • Step 305 Generate, according to the obtained behavior record, a popularity of each item or information in the second network platform within a preset time period.
  • the popularity of each item or information in the second network platform may be generated according to the behavior records, where the popularity may be specifically the item or the information within a preset time period.
  • the method for generating the popularity of the item or the information may be set according to actual conditions. For example, when a user purchases an item on the second network platform, the popularity of the item may be increased by one, or may be a After the user browses and collects an item on the second network platform, the popularity of the item may also be increased by one. When a user browses a piece of information on the second network platform, the popularity of the information may be increased by one. In this way, the popularity of each item or information is generated.
  • the popularity also changes with the length of time. If the preset time period is short, the popularity of the item or information is low. If the preset time period is long, the popularity of the item or information is quite different.
  • Step 306 Approve an item or information in the preset time period that has a popularity greater than a preset popularity threshold to each user in the social circle that is not in contact with the item or information.
  • each item or information may also be sorted in descending order, and the items or information with the highest popularity ranked directly to the users in the social circle who are not in contact with the item or information. Since the hobbies or interests of users in a social circle are similar, items or information that are more popular in the social circle are generally welcomed by users in the social circle.
  • 4 is a schematic diagram of a social circle-based collaborative recommendation process in the second embodiment of the information recommendation method of the present invention. As shown in FIG. 4, an item or information popular in a social circle is recommended to the social circle. Other users of the item or information, for example, user A and user B in a social circle like and pay attention to item 1, then item 1 can be recommended to user C in the social group.
  • an interface opened by an operator or a social network service provider includes a user identity acquisition interface, a friend relationship interface, a user behavior data interface, and a user registration.
  • Information interface from which social data is obtained, including user interaction information, user behavior information, and user identity.
  • the recommendation engine performs social network analysis, such as buddy extraction (ie, dividing friends ⁇ ), social ⁇ extraction (ie, dividing social ⁇ ), and calculating personal preference attributes through user behavior records and item or information records saved locally on the e-commerce website. , social ⁇ dice preference attributes.
  • the recommendation engine then performs specific information recommendation through the content recommendation policy and/or the collaborative recommendation strategy, and finally displays the recommendation result to the user through the portal.
  • the present embodiment provides an information recommendation method, which acquires relationship data information of users associated with users in the second network platform through an open interface of the first network platform, and acquires a second network platform according to relationship data information of each user.
  • the social user of each user divides each user and each user's social user into a friend circle corresponding to each user, and divides each friend circle into a plurality of different social circles according to the relationship data information, according to the user in the second
  • the behavior record in the network platform adopts the collaborative recommendation strategy to perform information recommendation in the divided social circle.
  • This embodiment can utilize the open interface and user data of the social website to perform information recommendation, improve the accuracy of information recommendation, and provide the user with the accuracy. Great convenience.
  • FIG. 6 is a flowchart of Embodiment 3 of the information recommendation method of the present invention. As shown in FIG. 6, the embodiment provides an information recommendation method, which may specifically include the following steps:
  • Step 601 The relationship data of the user associated with each user in the second network platform is obtained through the open interface of the first network platform. This step may be similar to the foregoing step 301, and details are not described herein again.
  • Step 602 Acquire, according to the relationship data information of each user, the social users of the users in the second network platform, and divide the social users of the users and the users into the friends corresponding to the users respectively. It can be similar to the above step 302, and details are not described herein again.
  • Step 603 Corresponding to each user according to the relationship data information of the user You can divide the Friend into a number of different social circles. This step can be similar to Step 303 above, and details are not described here.
  • Step 604 Obtain a behavior record of each user in the social circle in the second network platform.
  • the present embodiment performs information recommendation based on each social network.
  • the content recommendation policy and/or the collaboration recommendation policy may be used for the recommendation.
  • the content recommendation policy is taken as an example for description.
  • the specific collaboration recommendation policy refer to the second embodiment.
  • the item or information obtained by adopting the collaborative recommendation strategy is recommended to the user in the same social circle, and the item or information obtained by using the content recommendation strategy is also recommended to the user in the same social network.
  • This step is described by taking the information recommendation process in a social circle as an example.
  • the behavior record of each user in the social circle in the second network platform is obtained, where the behavior record includes the purchase record of the item and the browsing record of the information. .
  • Step 605 Calculate personal preference attributes of each user according to behavior records and relationship data information of each user, and use a common personal preference attribute of each user in the social circle as a circle preference attribute of the social circle.
  • the personal preference attributes of each user are respectively calculated according to the behavior record and the relationship data information of each user.
  • a user's preferences can be multifaceted, such as a user can discuss technical issues in a field in a technical circle, discuss the activity route of an event in an outdoor activity, and discuss child education in the family circle. Problems, etc.
  • the embodiment is based on user interaction information such as discussion and communication between the user and the friend participating on the first network platform, user behavior information such as microblogs and blogs published by the user on the second network platform, and the user is on the second network platform.
  • the behavior record of the purchased item or the browsed information can infer the user's preference, that is, the user's personal preference attribute can be obtained.
  • the personal preference attributes of each user in a social circle can be respectively obtained, and then the common personal preference attribute of each user in the social circle is used as the circle preference attribute of the social circle.
  • Step 606 Calculate a matching degree of an attribute of each item or information in the second network platform with a dice preference attribute of the social network.
  • the second network platform After obtaining the circle preference attribute of a social circle, the second network platform can be calculated.
  • the degree to which the attribute of the item or information matches the social preference's dice preference attribute, wherein the attribute of the item or information may be obtained according to the classification or characteristics of the item or information.
  • the attribute of the item or the information and the dice preference attribute of the social ⁇ may each be represented by a vector, and the vector includes a feature item describing the attribute, and then the calculation is performed.
  • D Document
  • T the feature item
  • the vector can be represented as D(a, b, c, d) o.
  • DDH D(WW 2 , ..., W n ).
  • Step 607 Approve an item or information whose matching degree is greater than a preset matching degree threshold to each user in the social network.
  • FIG. 7 is a schematic diagram of a content recommendation process based on a social network in the third embodiment of the information recommendation method of the present invention. As shown in FIG. 7, an item or information matching a circle preference attribute of the social circle is recommended to the social network. Each user.
  • the embodiment provides an information recommendation method, and acquires, by using an open interface of the first network platform, relationship data of a user associated with each user in the second network platform, according to each user.
  • the relationship data information is respectively obtained by the social users of the users in the second network platform, and the social users of the users and the users are respectively divided into the circle of friends corresponding to each user, and each circle of friends is divided into a plurality of different groups according to the relationship data information.
  • the social circle according to the behavior record of the user in the second network platform, adopts the content recommendation strategy to perform information recommendation in the divided social circle; the embodiment can utilize the open interface of the social website and the user data for information recommendation, and improves the The accuracy of information recommendation provides great convenience for users.
  • the aforementioned program can be stored in a computer readable storage medium.
  • the program when executed, performs the steps including the above-described method embodiments; and the foregoing storage medium includes: a medium that can store program codes, such as a ROM, a RAM, a magnetic disk, or an optical disk.
  • FIG. 8 is a schematic structural diagram of Embodiment 1 of the information recommendation apparatus according to the present invention.
  • the embodiment provides an information recommendation apparatus, which may specifically include the steps in the first embodiment of the foregoing method. Narration.
  • the information recommendation apparatus provided in this embodiment may specifically include an obtaining module 801, a dividing module 802, and a recommending module 803.
  • the obtaining module 801 is configured to acquire, by using an open interface of the first network platform, relationship data information of a user associated with each user in the second network platform, where the relationship data information includes user interaction information and interactions between the users.
  • User behavior information that represents the user's own behavior.
  • the dividing module 802 is configured to divide each friend circle divided according to the preset dividing policy according to the relationship data information of the user, and divide one friend ⁇ into a plurality of different social ⁇ .
  • the recommendation module 803 is configured to perform information recommendation in each social circle according to the obtained behavior record of each user in the second network platform, using a preset recommendation policy.
  • FIG. 9 is a schematic structural diagram of Embodiment 2 of the information recommendation apparatus according to the present invention.
  • this embodiment provides an information recommendation apparatus, which may specifically include the steps in the second embodiment of the foregoing method. Narration.
  • the information recommendation apparatus provided in this embodiment is based on the foregoing FIG. 8 , and the division module 802 may specifically include a first acquisition unit 812 , a first division unit 822 , and a second division unit 832 .
  • the first obtaining unit 812 is configured to acquire, according to the relationship data information of each user, the social users of the users in the second network platform, where the social users of the users are users who have social relationships with the users.
  • the first dividing unit 822 is configured to separately divide the social users of the users and the users into the corresponding users. Friends.
  • the second dividing unit 832 is configured to separately divide the buddy map corresponding to each user according to the relationship data information of the user, and divide one buddy circle into a plurality of different social circles.
  • the recommendation module 803 in this embodiment may be specifically configured to use a collaborative recommendation policy and/or a content recommendation policy in each of the social networks according to the obtained behavior record of each user in the second network platform. Make information recommendations.
  • the recommendation module 803 in this embodiment may specifically include a second obtaining unit 813, a generating unit 823, and a first recommending unit 833.
  • the second obtaining unit 813 is configured to obtain a behavior record of each user in the social network in the second network platform, where the behavior record includes a purchase record of the item and a browsing record of the information.
  • the generating unit 823 is configured to generate, according to the acquired behavior record, the popularity of each item or information in the second network platform within a preset time period. The item or information is recommended to each user in the social circle who is not in contact with the item or information.
  • the recommendation module 803 in this embodiment may specifically include a third obtaining unit 843, a first calculating unit 853, a second calculating unit 863, and a second recommending unit 873.
  • the third obtaining unit 843 is configured to obtain a behavior record of each user in the social network in the second network platform, where the behavior record includes a purchase record of the item and a browsing record of the information.
  • the first calculating unit 853 is configured to separately calculate personal preference attributes of each user according to behavior records and relationship data information of each user, and use a common personal preference attribute of each user in the social circle as a dice of the social circle. Preferences attribute.
  • the second calculating unit 863 is configured to calculate a matching degree of an attribute of each item or information in the second network platform with a circle preference attribute of the social circle.
  • the second recommendation unit 873 is configured to recommend items or information whose matching degree is greater than a preset matching degree threshold to each user in the social circle.
  • the embodiment provides an information recommendation device, which acquires relationship data information of a user associated with each user in the second network platform through an open interface of the first network platform, and acquires a second network platform according to the relationship data information of each user.
  • the social user of each user divides each user and each user's social user into a friend circle corresponding to each user, and divides each friend circle into a plurality of different social circles according to the relationship data information, according to the user in the second
  • the behavior record in the network platform adopts a preset recommendation strategy to perform information recommendation in the divided social circle.
  • This embodiment can utilize the open interface of the social website and user data for information recommendation, and improve information push. The accuracy of the recommendation provides great convenience to the user.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Strategic Management (AREA)
  • Marketing (AREA)
  • Accounting & Taxation (AREA)
  • Development Economics (AREA)
  • General Business, Economics & Management (AREA)
  • Finance (AREA)
  • Economics (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Game Theory and Decision Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Computing Systems (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Human Resources & Organizations (AREA)
  • Primary Health Care (AREA)
  • Tourism & Hospitality (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Information Transfer Between Computers (AREA)

Abstract

本发明实施例提供一种信息推荐方法和装置,方法包括:通过第一网络平台的开放接口获取与第二网络平台中的各用户相关联的用户的关系数据信息,关系数据信息包括各用户之间交互的用户交互信息和表示用户自身行为的用户行为信息;根据用户的关系数据信息分别对根据预设的划分策略划分得到的各好友圈进行划分,将一个好友圈划分为多个不同的社交圈;根据获取的各用户在第二网络平台中的行为记录,釆用预设的推荐策略在各所述社交圈中分别进行信息推荐。本发明实施例还提供了一种信息推荐装置。本实施例能够利用社交网站开放的接口和用户数据进行信息推荐,提高了信息推荐的准确度。

Description

信息推荐方法和装置
技术领域
本发明涉及互联网技术领域, 尤其涉及一种信息推荐方法和装置。 背景技术
在电子商务领域中, 随着电子商务规模的不断扩大, 商品个数和种类快 速增长, 顾客需要花费大量的时间才能找到自己想买的商品, 这种浏览大量 无关信息和产品的过程无疑会使淹没在信息过载问题中的消费者不断流失。 在互联网领域中, 随着博客、 维基、 微博的发展, 大量的网络信息由用户个 人产生, 信息组织散乱, 质量和可信度参差不齐, 使得用户需要花费大量时 间才能找到自己感兴趣的信息。 为了解决上述问题, 个性化推荐技术和个性 化推荐引擎应运而生。 个性化推荐技术是互联网领域, 特别是电子商务中的 重要技术, 其能根据用户的兴趣特点和购买能力, 向用户推荐用户感兴趣的 信息和商品。个性化推荐引擎是建立在海量数据挖掘基 上的一种智能平台, 以帮助电子商务网站、 互联网信息供应网站为其用户提供完全个性化的决策 支持和信息服务。
当前最主要的个性化推荐技术为基于内容的推荐和协同推荐。 基于内容 的推荐是指根据推荐物品或内容的元数据, 发现物品或信息的相关性, 给用 户推荐与其历史兴趣相关的物品或信息。 例如, 电子商务网站通过用户购买 记录发现, 用户 A在历史上总购买历史类书籍, 且用户 A还未购买当前很畅 销的历史书 "物品 3" , 因此推测用户 A为 "物品 3" 的潜在用户, 则将 "物 品 3" 推荐给用户 A。 协同推荐是指通过用户的历史行为记录发现用户的相 关性, 根据与用户相关的其他用户的兴趣做出的推荐。 例如, 电子商务网站 通过用户购买记录发现, 用户 A和用户 C在历史上总是购买相同的商品, 因 此推断用户 A和用户 C的兴趣爱好相似; 通过用户购买记录还发现用户 A 购买过 "物品 1" , 而用户 C尚未购买, 因此推测用户 C是 "物品 1" 的潜 在用户, 则将 "物品 1" 推荐给用户 C。
然而, 现有技术的推荐方法只适用于利用电子商务网站自身的用户数据 和历史数据进行推荐的场景, 信息推荐的准确度较低。 发明内容
本发明实施例提供一种信息推荐方法和装置, 能够利用社交网站开放 的接口和用户数据进行信息推荐, 提高信息推荐的准确度, 为用户提供极 大便利。
本发明实施例的第一个方面是提供一种信息推荐方法, 包括: 通过第一网络平台的开放接口获取与第二网络平台中的各用户相关 联的用户的关系数据信息, 所述关系数据信息包括各用户之间交互的用户 交互信息和表示用户自身行为的用户 4亍为信息;
根据所述用户的关系数据信息分别对根据预设的划分策略划分得到 的各好友圈进行划分, 将一个好友圈划分为多个不同的社交圈;
根据获取的各用户在所述第二网络平台中的行为记录, 采用预设的推 荐策略在各所述社交圏中分别进行信息推荐。
本发明实施例的另一个方面是提供一种信息推荐装置, 包括: 获取模块, 用于通过第一网络平台的开放接口获取与第二网络平台中 的各用户相关联的用户的关系数据信息, 所述关系数据信息包括各用户之 间交互的用户交互信息和表示用户自身行为的用户行为信息;
划分模块, 用于根据所述用户的关系数据信息分别对根据预设的划分 策略划分得到的各好友圈进行划分, 将一个好友圈划分为多个不同的社交 圏;
推荐模块, 用于根据获取的各用户在所述第二网络平台中的行为记 录, 采用预设的推荐策略在各所述社交圈中分别进行信息推荐。
本发明实施例的技术效果是: 通过第一网络平台的开放接口获取与第 二网络平台中各用户相关联的用户的关系数据信息, 根据该关系数据信息 分别将各好友圈划分为多个不同的社交圈, 根据用户在第二网络平台中的 行为记录, 在划分后的社交圈中进行信息推荐; 本实施例能够利用社交网 站开放的接口和用户数据进行信息推荐, 提高了信息推荐的准确度, 为用 户提供了极大便利。 附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案, 下面将对 实施例描述中所需要使用的附图作一简单地介绍, 显而易见地, 下面描述 中的附图是本发明的一些实施例, 对于本领域普通技术人员来讲, 在不付 出创造性劳动性的前提下, 还可以根据这些附图获得其他的附图。
图 1为本发明信息推荐方法实施例一的流程图;
图 2为本发明信息推荐方法实施例一中好友圈与社交圈的关系示意 图;
图 3为本发明信息推荐方法实施例二的流程图;
图 4为本发明信息推荐方法实施例二中基于社交圈的协同推荐过程的 示意图;
图 5为本发明信息推荐方法实施例二中的系统架构示意图; 图 6为本发明信息推荐方法实施例三的流程图;
图 7为本发明信息推荐方法实施例三中基于社交圈的内容推荐过程的 示意图;
图 8为本发明信息推荐装置实施例一的结构示意图;
图 9为本发明信息推荐装置实施例二的结构示意图。 具体实施方式
为使本发明实施例的目的、 技术方案和优点更加清楚, 下面将结合本 发明实施例中的附图, 对本发明实施例中的技术方案进行清楚、 完整地描 述, 显然, 所描述的实施例是本发明一部分实施例, 而不是全部的实施例。 基于本发明中的实施例, 本领域普通技术人员在没有作出创造性劳动前提 下所获得的所有其他实施例, 都属于本发明保护的范围。
针对现有技术的信息推荐方案, 电子商务网站利用自身产生的历史数 据进行推荐时, 历史数据和潜在用户信息、 待推荐物品的信息之间具有高 度的一致性, 潜在用户就是本网站购买过物品的用户, 待推荐物品与历史 数据中用户购买过的物品相同或非常相似。 因此, 需要结合运营商网站或 社交网站产生的历史数据来进行电子商务网站的信息推荐, 以克服上述推 荐的物品或信息对用户而言价值不高的缺陷, 而传统的基于电子商务网站 自身产生的历史数据进行推荐的方法, 无法直接适用于基于运营商网站或 社交网站产生的社交数据进行推荐。 由于近年来互联网业界的趋势发展发 生了变化, 各个运营商或社交网站服务商愿意将自身的各种资源以应用程 序编程接口 ( Application Programming Interface; 以下简称: API ) 的形式 开放出来, 以将自身打造成一个开放平台, 吸引开发者在自己的平台上开 发增值业务。 因此, 业界的一个新需求便是如何利用运营商或社交网站服 务商开放出来的接口和用户数据, 进行物品或信息推荐。 本发明旨在解决 上述技术问题, 提出一种信息推荐方法, 利用运营商或社交网站服务商开 放的资源, 进行物品或信息的准确推荐。
图 1为本发明信息推荐方法实施例一的流程图, 如图 1所示, 本实施 例提供了一种信息推荐方法, 可以具体包括如下步骤:
步骤 101 , 通过第一网络平台的开放接口获取与第二网络平台中的各 用户相关联的用户的关系数据信息。
本实施例中的第一网络平台可以具体为运营商网站或社交网站, 如 facebook、 twitter, 新浪微博等, 第二网络平台可以具体为电子商务网站, 如淘宝网、 京东商城、 当当网等。 在本实施例中, 第一网络平台向外开放 了 API接口, 第二网络平台可以通过第一网络平台的开放接口获取用户的 关系数据信息。 此处的用户为与第二网络平台中的用户相关联的用户, 此 处的相关联是指两个网络平台上具有相同身份信息的用户, 例如同一个人 使用相同或不同的账号在第一网络平台和第二网络平台进行注册, 成为两 个网络平台的用户。 虽然第一网络平台和第二网络平台是两个独立的平 台, 各自有自己的用户, 但由于第一网络平台对外开放了接口, 第二网络 平台可以通过开放接口找到一种两个平台的用户相关联的方法, 即第二网 络平台可以通过第一网络平台开放的用户的注册信息, 如邮箱地址等, 从 注册信息中识别用户的身份, 再通过第二网络平台自身的用户的注册信息 识別第二网络平台中用户的身份。 如果两个网络平台中两个用户的身份相 同, 则这两个用户为相关联的用户。 此处的关系数据信息包括各用户之间 交互的用户交互信息和表示用户 自身行为的用户行为信息, 用户交互信息 可以为第一网络平台中用户与好友之间的相互发送电子邮件、 短消息的行 为, 或者博客、 微博中好友之间的相互浏览、 转发、 评论等行为; 以及这 些行为相关的文本数据, 可以用于对用户的好友圈进行进一步划分。 用户 行为信息可以为用户自己发表的博客、 微博等信息, 可以用于确定用户的 个人偏好属性。
步骤 102, 根据所述用户的关系数据信息分别对根据预设的划分策略 划分得到的各好友圏进行划分, 将一个好友圈划分为多个不同的社交圈。
在获取到用户的关系数据信息后, 可以根据该关系数据信息分别对各 好友圈进行划分, 具体将一个好友圏划分为多个不同的社交圈。 此处的好 友圈为根据预设的划分策略划分得到的, 划分策略可以为以用户为中心的 划分策略, 也可以为按照网络的聚集度进行划分的划分策略, 通过预设的 划分策略可以将网络中的用户划分为多个好友圏, 不同好友圏之间通常可 能包括一个或几个相同的用户, 即不同好友圈之间存在相互重叠的情况。 本步骤根据从第一网络平台获取到的用户交互信息和用户行为信息, 分别 对各好友圏进行进一步的划分, 即通过好友之间的互动情况以及用户发表 或参与的交流讨论话题, 可以确定用户与其好友的关系, 如同学、 同事、 家人, 某主题的学术圏或交流讨论圏等, 从而将一个好友圏可以划分为多 个不同的社交圏。 图 2为本发明信息推荐方法实施例一中好友圈与社交圏 的关系示意图, 如图 2所示, 将一个用户的好友圈划分为四个社交圈, 分 别为技术圏、 同事圏、 家人圈和户外活动圈。
步骤 103 , 根据获取的各用户在所述第二网络平台中的行为记录, 采 用预设的推荐策略在各所述社交圈中分别进行信息推荐。
在完成社交圈的划分之后, 本步骤采用预设的推荐策略在各社交圈中 分别进行信息推荐, 此处的推荐策略可以具体为协同推荐策略, 或内容推 荐策略, 或协同推荐策略和内容推荐策略的结合。 本实施例具体以一个社 交圈为单位, 根据获取的该社交圈中各用户在第二网络平台中的行为记 录, 采用预设的推荐策略进行信息推荐。 此处的用户在第二网络平台中的 行为记录可以包括用户在第二网络平台中的物品的购买记录以及信息的 浏览记录等。 由于一个社交圈中各用户的爱好相似, 其所关注或关心的话 题类似, 因此, 基于该社交圈来推荐其中流行度较高的物品或信息, 该圈 子中的其他用户通常会对推荐的物品或信息感兴趣, 从而提高了推荐的准 确度, 同时用户无需盲目搜索便可以获取到其感兴趣的物品或信息, 也为 用户提供了便利。
本实施例提供了一种信息推荐方法, 通过第一网络平台的开放接口获 取与第二网络平台中各用户相关联的用户的关系数据信息, 根据该关系数 据信息分别将各好友圏划分为多个不同的社交圈, 根据用户在第二网络平 台中的行为记录, 在划分后的社交圏中进行信息推荐; 本实施例能够利用 社交网站开放的接口和用户数据进行信息推荐, 提高了信息推荐的准确 度, 为用户提供了极大便利。
图 3为本发明信息推荐方法实施例二的流程图, 如图 3所示, 本实施 例提供了一种信息推荐方法, 可以具体包括如下步骤:
步骤 301 , 通过第一网络平台的开放接口获取与第二网络平台中的各 用户相关联的用户的关系数据信息。
在本实施例中, 第一网络平台向外开放了 API接口, 第二网络平台可 以通过第一网络平台的开放接口获取用户的关系数据信息。 此处的用户为 与第二网络平台中的用户相关联的用户, 此处的相关联是指两个网络平台 上具有相同身份信息的用户。 虽然第一网络平台和第二网络平台是两个独 立的平台, 各自有自己的用户, 但由于第一网络平台对外开放了接口, 第 二网络平台可以通过开放接口找到一种两个平台的用户相关联的方法, 即 第二网络平台可以通过第一网络平台开放的用户的注册信息, 从注册信息 中识别用户的身份, 再通过第二网络平台自身的用户的注册信息识别第二 网络平台中用户的身份。 如果两个网络平台中两个用户的身份相同, 则这 两个用户为相关联的用户。
步骤 302, 根据各用户的关系数据信息分别获取第二网络平台中各用 友圈。
在现有技术中, 将历史上购买相同物品的用户看作是相似用户, 一个 用户购买某一个物品后, 可以认为该用户的相似用户为该物品的潜在客 户。 然而, 实际应用中发现, 现有技术中这种潜在客户的识别方法精确度 不高, 容易对用户造成推荐干扰, 即向用户推荐其并不感兴趣的物品或信 息, 如果这种现象频繁的话会对用户造成一定干扰。 为了克服现有技术中 的上述推荐精确度不高的缺陷, 本实施例通过对从第一网络平台获取的用 户的关系数据信息进行分析, 进而准确地识别潜在用户。 社交网络中用户 之间是一个巨大的关系网络, 在识别潜在用户时, 需要根据该网络的拓朴 结构进行分割形成多个小的子网, 此处的一个子网可以为一个好友圈。 本 实施例先根据上述步骤获取到的各用户的关系数据信息, 分别获取第二网 络平台中各用户的社交用户, 此处的用户的社交用户为与各用户具有社交 关系的用户, 社交关系具体指用户之间通过第一网络平台进行的问题交 流、 相互评论、 转发微博等。 本步骤将用户和该用户的社交用户划分为该 用户对应的好友圈, 即以某一个用户为中心, 将与该用户具有社交关系的 其他用户与该用户一起组成一个好友圈, 该好友圏具体为该用户的好友 圏; 还可以以另外一个用户为中心, 建立另外一个用户对应的好友圏。 各 用户对应的好友圈各不相同, 但不同好友圏之间可能存在重叠部分, 即具 有共同的好友, 如图 4所示为建立的一个好友圈。 具体地, 一个好友圈中 可能包含多层好友关系, 例如两层好友关系为: 假设以用户 A为中心, 用 户 B为用户 A的好友, 用户 C为用户 B的好友, 则将用户 C也加入到用 户 A对应的好友圏中。
或者, 本实施例也可以按照社交网络的聚集度来划分形成好友圈, 即 可以将社交网络中相互连接紧密的节点形成一个子网, 该子网即为一个好 友圈。 此处的社交网络可以为根据用户之间的关系形成的一个网络, 网络 中的每个节点即代表每个用户, 网络中两个节点相互连接表示这两个用户 之间存在交互行为, 如相互浏览、 转发微博等的行为。
步骤 303 , 根据用户的关系数据信息分别对各用户对应的好友圈进行 划分, 将一个好友圈划分为多个不同的社交圈。
本步驟为在划分得到好友圈后, 由于每个好友圈涉及的用户群太广, 则需要进一步对用户的好友进行筛选, 以更准确地识别潜在客户。 具体为 根据从第一网络平台获取到的用户交互信息和用户行为信息, 分别对各好 友圏进行进一步的划分, 即通过好友之间的互动情况以及用户发表或参与 的交流讨论话题, 可以确定用户与其好友的关系, 如同学、 同事、 家人, 某主题的学术圈或交流讨论圈等, 从而将一个好友圈可以划分为多个不同 的社交圈。 如图 2所示, 将一个用户的好友圈划分为四个社交圏, 分別为 技术圈、 同事圈、 家人圈和户外活动圏, 划分后的每个社交圈中的用户便 可以当成某类或某个商品或信息的潜在客户。
步骤 304, 获取一个社交圈中各用户在所述第二网络平台中的行为记 录。
在对各好友圏进行划分得到各自的社交圈后, 本实施例基于每个社交 圏进行信息推荐。 具体可以采用内容推荐策略和 /或协同推荐策略进行推 荐, 本实施例以协同推荐策略为例进行说明。 本步骤以在一个社交圈中的 信息推荐过程为例进行说明, 先获取一个社交圈中各用户在第二网络平台 中的行为记录, 此处的行为记录包括物品的购买记录和信息的浏览记录。
步骤 305, 根据获取的行为记录生成所述第二网络平台中各物品或信 息在预设时间段内的流行度。
在获取到该社交圏中各用户的行为记录后, 可以根据这些行为记录生 成第二网絡平台中各物品或信息的流行度, 此处的流行度可以具体为物品 或信息在预设时间段内的流行度。 物品或信息的流行度的生成方法可以根 据实际情况来设定,例如, 当一个用户在第二网络平台上购买一个物品后, 则可以将该物品的流行度加 1 , 或者也可以为当一个用户在第二网络平台 上浏览并收藏一个物品后, 也可以将该物品的流行度加 1 , 当一个用户在 第二网络平台上浏览一条信息后, 则可以将该信息的流行度加 1, 以此来 生成各物品或信息的流行度。 物品或信息的流行度越高, 表明该物品或信 息在第二网络平台中越受欢迎, 当然, 此处的流行度具体与一个社交圈相 对应。 流行度也随时间的长短而发生变化, 如果预设时间段较短, 则物品 或信息的流行度均较低, 如果预设时间段较长, 则物品或信息的流行度的 差异较大。
步骤 306 , 将所述预设时间段内流行度大于预设的流行度阈值的物品 或信息推荐给所述社交圈中未接触该物品或信息的各用户。
在生成第二网络平台上物品或信息在预设时间段内的流行度后, 将该 预设时间段内流行度大于预设的流行度阈值的物品或信息推荐给该社交 圏中的各用户, 也可以对各物品或信息的流行度按照从大到小的顺序进行 排序, 将流行度排在前几位的物品或信息直接推荐给该社交圈中未接触该 物品或信息的各用户。 由于一个社交圈中各用户之间的爱好或兴趣类似, 则在该社交圈中流行度较高的物品或信息通常都受社交圏中的用户欢迎。 图 4为本发明信息推荐方法实施例二中基于社交圈的协同推荐过程的示意 图, 如图 4所示, 将某个社交圈中流行的物品或信息, 推荐给该社交圈中 未接触过该物品或信息的其他用户, 例如, 某个社交圈中用户 A和用户 B 均喜欢并关注了物品 1, 则可以将该物品 1推荐给该社交圏中的用户 C。
图 5为本发明信息推荐方法实施例二中的系统架构示意图, 如图 5所 示, 运营商或社交网络服务商开放的接口包括用户身份获取接口、 好友关 系接口、 用户行为数据接口、 用户注册信息接口, 从这些接口获取社交数 据, 包括用户交互信息、 用户行为信息以及用户身份。 然后, 推荐引擎通 过电子商务网站本地保存的用户行为记录以及物品或信息记录, 进行社会 化网络分析 ,如好友提取(即划分好友圏)、社交圏提取(即划分社交圏), 计算个人偏好属性、 社交圏的圏子偏好属性。 推荐引擎再通过内容推荐策 略和 /或协同推荐策略进行具体的信息推荐,最终通过 Portal将推荐结果显 示给用户。
本实施例提供了一种信息推荐方法, 通过第一网络平台的开放接口获 取与第二网络平台中各用户相关联的用户的关系数据信息, 根据各用户的 关系数据信息分别获取第二网络平台中各用户的社交用户, 将各用户和各 用户的社交用户分別划分为各用户对应的好友圈, 根据该关系数据信息分 别将各好友圈划分为多个不同的社交圈, 根据用户在第二网络平台中的行 为记录, 采用协同推荐策略在划分后的社交圈中进行信息推荐; 本实施例 能够利用社交网站开放的接口和用户数据进行信息推荐, 提高了信息推荐 的准确度, 为用户提供了极大便利。
图 6为本发明信息推荐方法实施例三的流程图, 如图 6所示, 本实施 例提供了一种信息推荐方法, 可以具体包括如下步骤:
步骤 601 , 通过第一网络平台的开放接口获取与第二网络平台中的各 用户相关联的用户的关系数据信息, 本步骤可以与上述步驟 301类似, 此 处不再赘述。
步骤 602, 根据各用户的关系数据信息分别获取所述第二网络平台中 各用户的社交用户, 将各用户和所述各用户的社交用户分别划分为所述各 用户对应的好友圏, 本步骤可以与上述步骤 302类似, 此处不再赘述。
步骤 603 , 根据所述用户的关系数据信息分别对所述各用户对应的好 友圏进行划分, 将一个好友圏划分为多个不同的社交圈, 本步驟可以与上 述步骤 303类似, 此处不再赘述。
步骤 604, 获取一个社交圈中各用户在所述第二网络平台中的行为记 录。
在对各好友圏进行划分得到各自的社交圈后, 本实施例基于每个社交 圏进行信息推荐。 具体可以采用内容推荐策略和 /或协同推荐策略进行推 荐, 本实施例以内容推荐策略为例进行说明, 具体协同推荐策略可以参见 上述实施例二; 对于协同推荐策略和内容推荐策略相结合的方案, 则为将 采用协同推荐策略获得的物品或信息推荐给同一社交圈中的用户, 同时将 采用内容推荐策略获得的物品或信息也推荐给同一社交圏中的用户。 本步 骤以在一个社交圈中的信息推荐过程为例进行说明, 先获取一个社交圈中 各用户在第二网络平台中的行为记录, 此处的行为记录包括物品的购买记 录和信息的浏览记录。
步骤 605, 根据各用户的行为记录和关系数据信息分别计算所述各用 户的个人偏好属性, 将社交圈中各用户的共同的个人偏好属性作为所述社 交圈的圈子偏好属性。
在获取到社交圏中各用户的行为记录和各用户的关系数据信息后, 根 据各用户的行为记录和关系数据信息分别计算各用户的个人偏好属性。一 个用户的偏好可以时多方面的, 如一个用户可以在一个技术圈讨论某领域 的技术问题, 也可以在一个户外活动圏讨论某次活动的活动路线, 还可以 在家庭圈中讨论孩子的教育问题等。 本实施例基于用户在第一网络平台上 参与的与其好友之间的讨论、 交流等用户互动信息、 用户在第二网络平台 上发表的微博、 博客等用户行为信息以及用户在第二网络平台上购买的物 品或浏览的信息等行为记录, 可以推断出该用户的爱好, 即可以获取到该 用户的个人偏好属性。依照上述方法可以分别获取到一个社交圈中各用户 的个人偏好属性, 然后将该社交圏中各用户的共同的个人偏好属性作为该 社交圈的圈子偏好属性。
步骤 606, 计算所述第二网络平台中各物品或信息的属性与所述社交 圏的圏子偏好属性的匹配程度。
在获取到某个社交圈的圈子偏好属性后, 可以计算第二网络平台中各 物品或信息的属性与该社交圏的圏子偏好属性的匹配程度, 其中, 物品或 信息的属性可以为根据物品或信息的分类、 特点获取得到。
在计算物品或信息的属性与圈子偏好属性的匹配程度时, 可以将物品 或信息的属性和社交圏的圏子偏好属性各用一个向量表示, 向量中包含有 描述属性的特征项, 然后计算这两个向量的相关度。 在向量空间模型中, 用 D ( Document )表示向量, 特征项 (Term, 用 T表示)是指向量 D中 的特征项, 向量可以用特征项集表示为 Dd , Τ2, ..., Τη), 其中 Tk是特 征项, l<=k<=N。 例如一个向量中有 a、 b、 c、 d四个特征项, 那么这个 向量就可以表示为 D(a, b, c , d) o 对含有 n个特征项的向量而言, 通常 会给每个特征项赋予一定的权重表示其重要程度。 即 D D H , Wi ; T2 , W2 ; ..., Tn, Wn), 简记为 D = D(W W2, ..., Wn)。 其中 Wk Tk的权 重, l<=k<=N。 在上面那个例子中, 假设 a、 b、 c、 d的权重分别为 30 , 20 , 20 , 10 , 那么该文本的向量表示为 D(30 , 20 , 20 , 10)。 在向量空间 模型中, 两个文档 01和1)2之间的相关度 Sim D2)常用向量之间夹角 的余弦值表示, 如下迷公式 ( 1 ) 所示:
y A-J. ( 1 ) 其中, Wlk、 W2k分别表示文档 和 D2的第 k个特征项的权值, l<=k<=N。
步骤 607 , 将匹配程度大于预设的匹配程度阈值的物品或信息推荐给 所述社交圏中的各用户。
在获取到物品或信息的属性与社交圈的圈子偏好属性的匹配程度后, 将该匹配程度大于预设的匹配程度阈值的物品或信息推荐给该社交圏中 的各用户, 即将二者匹配程度较高的物品或信息向该社交圏中的各用户推 荐。 图 7为本发明信息推荐方法实施例三中基于社交圏的内容推荐过程的 示意图, 如图 7所示, 将与该社交圈的圈子偏好属性相匹配的物品或信息 推荐给该社交圏中的各用户。
本实施例提供了一种信息推荐方法, 通过第一网络平台的开放接口获 取与第二网络平台中各用户相关联的用户的关系数据信息, 根据各用户的 关系数据信息分别获取第二网络平台中各用户的社交用户, 将各用户和各 用户的社交用户分别划分为各用户对应的好友圈, 根据该关系数据信息分 别将各好友圈划分为多个不同的社交圈, 根据用户在第二网络平台中的行 为记录, 采用内容推荐策略在划分后的社交圈中进行信息推荐; 本实施例 能够利用社交网站开放的接口和用户数据进行信息推荐, 提高了信息推荐 的准确度, 为用户提供了极大便利。
本领域普通技术人员可以理解: 实现上迷各方法实施例的全部或部分 步骤可以通过程序指令相关的硬件来完成。 前述的程序可以存储于一计算 机可读取存储介质中。 该程序在执行时, 执行包括上述各方法实施例的步 骤; 而前述的存储介质包括: ROM、 RAM, 磁碟或者光盘等各种可以存 储程序代码的介质。
图 8为本发明信息推荐装置实施例一的结构示意图, 如图 8所示, 本 实施例提供了一种信息推荐装置, 可以具体包括执行上述方法实施例一中 的各个步骤, 此处不再赘述。 本实施例提供的信息推荐装置可以具体包括 获取模块 801、 划分模块 802和推荐模块 803。 其中, 获取模块 801用于 通过第一网络平台的开放接口获取与第二网络平台中的各用户相关联的 用户的关系数据信息, 所述关系数据信息包括各用户之间交互的用户交互 信息和表示用户自身行为的用户行为信息。 划分模块 802用于根据所述用 户的关系数据信息分别对根据预设的划分策略划分得到的各好友圈进行 划分, 将一个好友圏划分为多个不同的社交圏。 推荐模块 803用于根据获 取的各用户在所述第二网络平台中的行为记录, 釆用预设的推荐策略在各 所述社交圈中分别进行信息推荐。
图 9为本发明信息推荐装置实施例二的结构示意图, 如图 9所示, 本 实施例提供了一种信息推荐装置, 可以具体包括执行上述方法实施例二中 的各个步骤, 此处不再赘述。 本实施例提供的信息推荐装置在上述图 8所 示的基础之上, 划分模块 802可以具体包括第一获取单元 812、 第一划分 单元 822和第二划分单元 832。 其中, 第一获取单元 812用于根据各用户 的关系数据信息分别获取所述第二网络平台中各用户的社交用户, 所述各 用户的社交用户为与所迷各用户具有社交关系的用户。 第一划分单元 822 用于将所述各用户和所述各用户的社交用户分别划分为所述各用户对应 的好友圏。 第二划分单元 832用于根据所述用户的关系数据信息分別对所 述各用户对应的好友圖进行划分, 将一个好友圈划分为多个不同的社交 圈。
具体地, 本实施例中的推荐模块 803可以具体用于根据获取的各用户 在所述第二网络平台中的行为记录,采用协同推荐策略和 /或内容推荐策略 在各所述社交圏中分别进行信息推荐。
更具体地, 本实施例中的推荐模块 803可以具体包括第二获取单元 813、 生成单元 823和第一推荐单元 833。 其中, 第二获取单元 813用于获 取一个社交圏中各用户在所述第二网络平台中的行为记录, 所述行为记录 包括物品的购买记录和信息的浏览记录。 生成单元 823用于根据获取的行 为记录生成所述第二网络平台中各物品或信息在预设时间段内的流行度。 的物品或信息推荐给所述社交圈中未接触该物品或信息的各用户。
更具体地, 本实施例中的推荐模块 803可以具体包括第三获取单元 843、 第一计算单元 853、 第二计算单元 863和第二推荐单元 873。 其中, 第三获取单元 843用于获取一个社交圏中各用户在所述第二网络平台中的 行为记录, 所述行为记录包括物品的购买记录和信息的浏览记录。 第一计 算单元 853用于根据各用户的行为记录和关系数据信息分别计算所述各用 户的个人偏好属性, 将所述社交圈中各用户的共同的个人偏好属性作为所 述社交圈的圏子偏好属性。 第二计算单元 863用于计算所述第二网络平台 中各物品或信息的属性与所述社交圈的圈子偏好属性的匹配程度。 第二推 荐单元 873用于将匹配程度大于预设的匹配程度阈值的物品或信息推荐给 所述社交圈中的各用户。
本实施例提供了一种信息推荐装置, 通过第一网络平台的开放接口获 取与第二网络平台中各用户相关联的用户的关系数据信息, 根据各用户的 关系数据信息分别获取第二网络平台中各用户的社交用户, 将各用户和各 用户的社交用户分别划分为各用户对应的好友圈, 根据该关系数据信息分 别将各好友圈划分为多个不同的社交圈, 根据用户在第二网络平台中的行 为记录, 采用预设的推荐策略在划分后的社交圈中进行信息推荐; 本实施 例能够利用社交网站开放的接口和用户数据进行信息推荐, 提高了信息推 荐的准确度, 为用户提供了极大便利。
最后应说明的是: 以上各实施例仅用以说明本发明的技术方案, 而非 对其限制; 尽管参照前述各实施例对本发明进行了详细的说明, 本领域的 普通技术人员应当理解: 其依然可以对前述各实施例所记载的技术方案进 行修改, 或者对其中部分或者全部技术特征进行等同替换; 而这些修改或 者替换, 并不使相应技术方案的本质脱离本发明各实施例技术方案的范 围。

Claims

权 利 要 求 书
1、 一种信息推荐方法, 其特征在于, 包括:
通过第一网络平台的开放接口获取与第二网络平台中的各用户相关 联的用户的关系数据信息, 所述关系数据信息包括各用户之间交互的用户 交互信息和表示用户自身行为的用户行为信息;
根据所述用户的关系数据信息分别对根据预设的划分策略划分得到 的各好友圏进行划分, 将一个好友圈划分为多个不同的社交圏;
根据获取的各用户在所述第二网络平台中的行为记录, 釆用预设的推 荐策略在各所述社交圏中分别进行信息推荐。
2、 根据权利要求 1所述的方法, 其特征在于, 所述根据所述用户的关 系数据信息分别对根据预设的划分策略划分得到的各好友圈进行划分, 将 一个好友圈划分为多个不同的社交圈包括:
根据各用户的关系数据信息分别获取所述第二网络平台中各用户的 社交用户, 所述各用户的社交用户为与所述各用户具有社交关系的用户; 将所述各用户和所述各用户的社交用户分别划分为所述各用户对应 的好友圈;
根据所述用户的关系数据信息分別对所述各用户对应的好友圏进行 划分, 将一个好友圏划分为多个不同的社交圈。
3、 根据权利要求 1或 2所述的方法, 其特征在于, 所述根据获取的 各用户在所述第二网络平台中的行为记录, 采用预设的推荐策略在各所述 社交圏中分别进行信息推荐包括:
根据获取的各用户在所述第二网络平台中的行为记录, 采用协同推荐 策略和 /或内容推荐策略在各所迷社交圏中分别进行信息推荐。
4、 根据权利要求 3所述的方法, 根据获取的各用户在所述第二网络 平台中的行为记录, 采用协同推荐策略在各社交圈中分别进行信息推荐包 括:
获取一个社交圈中各用户在所述第二网络平台中的行为记录, 所述行 为记录包括物品的购买记录和信息的浏览记录;
根据获取的行为记录生成所述第二网络平台中各物品或信息在预设 时间段内的流行度; 将所述预设时间段内流行度大于预设的流行度阈值的物品或信息推 荐给所述社交圈中未接触该物品或信息的各用户。
5、 根据权利要求 3所述的方法, 其特征在于, 根据获取的各用户在 所述第二网络平台中的行为记录, 采用内容推荐策略在各所述社交圈中分 别进行信息推荐包括:
获取一个社交圏中各用户在所述第二网络平台中的行为记录, 所述行 为记录包括物品的购买记录和信息的浏览记录;
根据各用户的行为记录和关系数据信息分别计算所述各用户的个人 偏好属性;
将所述社交圏中各用户的共同的个人偏好属性作为所述社交圏的圏 子偏好属性;
计算所述第二网络平台中各物品或信息的属性与所述社交圏的圈子 偏好属性的匹配程度;
将匹配程度大于预设的匹配程度阈值的物品或信息推荐给所述社交 圏中的各用户。
6、 一种信息推荐装置, 其特征在于, 包括:
获取模块, 用于通过第一网络平台的开放接口获取与第二网络平台中 的各用户相关联的用户的关系数据信息, 所述关系数据信息包括各用户之 间交互的用户交互信息和表示用户自身行为的用户行为信息;
划分模块, 用于根据所述用户的关系数据信息分别对根据预设的划分 策略划分得到的各好友圈进行划分, 将一个好友圈划分为多个不同的社交 圏;
推荐模块, 用于根据获取的各用户在所述第二网络平台中的行为记 录, 采用预设的推荐策略在各所迷社交圈中分別进行信息推荐。
7、 根据权利要求 6所述的装置, 其特征在于, 所述划分模块包括: 第一获取单元, 用于根据各用户的关系数据信息分别获取所述第二网 络平台中各用户的社交用户, 所述各用户的社交用户为与所述各用户具有 社交关系的用户;
第一划分单元, 用于将所述各用户和所述各用户的社交用户分别划分 为所述各用户对应的好友圏; 第二划分单元, 用于根据所述用户的关系数据信息分別对所述各用户 对应的好友圈进行划分, 将一个好友圈划分为多个不同的社交圏。
8、 根据权利要求 6或 7所述的装置, 其特征在于, 所述推荐模块具 体用于根据获取的各用户在所述第二网络平台中的行为记录, 采用协同推 荐策略和 /或内容推荐策略在各所述社交圈中分别进行信息推荐。
9、 根据权利要求 8所述的装置, 其特征在于, 所述推荐模块包括: 第二获取单元, 用于获取一个社交圈中各用户在所迷第二网络平台中 的行为记录, 所述行为记录包括物品的购买记录和信息的浏览记录; 生成单元, 用于根据获取的行为记录生成所述第二网络平台中各物品 或信息在预设时间段内的流行度;
第一推荐单元, 用于将所述预设时间段内流行度大于预设的流行度阈 值的物品或信息推荐给所述社交圈中未接触该物品或信息的各用户。
10、 根据权利要求 8所述的装置, 其特征在于, 所述推荐模块包括: 第三获取单元, 用于获取一个社交圈中各用户在所述第二网络平台中 的行为记录, 所述行为记录包括物品的购买记录和信息的浏览记录; 第一计算单元, 用于根据各用户的行为记录和关系数据信息分别计算 所述各用户的个人偏好属性, 将所述社交圈中各用户的共同的个人偏好属 性作为所述社交圏的圏子偏好属性;
第二计算单元, 用于计算所述第二网络平台中各物品或信息的属性与 所述社交圏的圏子偏好属性的匹配程度;
第二推荐单元, 用于将匹配程度大于预设的匹配程度阈值的物品或信 息推荐给所述社交圈中的各用户。
PCT/CN2013/070158 2012-06-27 2013-01-07 信息推荐方法和装置 WO2014000431A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US14/333,784 US20140330653A1 (en) 2012-06-27 2014-07-17 Information Recommendation Method and Apparatus
US17/109,629 US20210133817A1 (en) 2012-06-27 2020-12-02 Information Recommendation Method and Apparatus

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201210215463.0A CN103514204B (zh) 2012-06-27 2012-06-27 信息推荐方法和装置
CN201210215463.0 2012-06-27

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US14/333,784 Continuation US20140330653A1 (en) 2012-06-27 2014-07-17 Information Recommendation Method and Apparatus

Publications (1)

Publication Number Publication Date
WO2014000431A1 true WO2014000431A1 (zh) 2014-01-03

Family

ID=49782165

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2013/070158 WO2014000431A1 (zh) 2012-06-27 2013-01-07 信息推荐方法和装置

Country Status (3)

Country Link
US (2) US20140330653A1 (zh)
CN (1) CN103514204B (zh)
WO (1) WO2014000431A1 (zh)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110737846A (zh) * 2019-10-28 2020-01-31 北京字节跳动网络技术有限公司 一种社交界面推荐方法、装置、电子设备及存储介质
CN112950325A (zh) * 2021-03-16 2021-06-11 山西大学 一种社交行为融合的自注意力序列推荐方法
US20220172260A1 (en) * 2020-02-07 2022-06-02 Tencent Technology (Shenzhen) Company Limited Method, apparatus, storage medium, and device for generating user profile
CN115348544A (zh) * 2021-05-14 2022-11-15 中国移动通信集团浙江有限公司 异网用户定位方法、终端设备及存储介质

Families Citing this family (22)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105095176A (zh) * 2014-04-29 2015-11-25 华为技术有限公司 用户设备提取文本信息中的特征信息的方法和用户设备
CN104052651B (zh) * 2014-06-03 2017-09-12 西安交通大学 一种建立社交组的方法和装置
CN108197330B (zh) 2014-11-10 2019-10-29 北京字节跳动网络技术有限公司 基于社交平台的数据挖掘方法及装置
CN105677162B (zh) * 2014-11-19 2019-12-03 深圳市腾讯计算机系统有限公司 匹配条件列表的显示方法和装置
WO2016095133A1 (en) * 2014-12-17 2016-06-23 Yahoo! Inc. Method and system for determining user interests based on a correspondence graph
CN104615775B (zh) * 2015-02-26 2018-08-07 北京奇艺世纪科技有限公司 一种用户推荐方法及装置
CN106302104B (zh) * 2015-06-26 2020-01-21 阿里巴巴集团控股有限公司 一种用户关系识别方法及装置
CN106301880B (zh) * 2015-06-29 2019-12-24 阿里巴巴集团控股有限公司 一种确定网络关系稳定度、互联网业务推荐方法和设备
CN106330846A (zh) * 2015-07-03 2017-01-11 阿里巴巴集团控股有限公司 跨平台的对象推荐方法及装置
CN111159540B (zh) * 2015-09-16 2024-03-08 那彦琳 数据处理方法及装置
CN106611350B (zh) * 2015-10-26 2020-06-05 阿里巴巴集团控股有限公司 挖掘潜在用户源的方法和装置
CN107545451B (zh) * 2016-06-27 2020-11-27 腾讯科技(深圳)有限公司 一种广告推送方法及装置
CN106383899A (zh) * 2016-09-28 2017-02-08 北京小米移动软件有限公司 服务推荐方法、装置及终端
CN111199491B (zh) * 2018-10-31 2023-08-22 百度在线网络技术(北京)有限公司 社交圈的推荐方法和装置
CN109615487A (zh) * 2019-01-04 2019-04-12 平安科技(深圳)有限公司 基于用户行为的产品推荐方法、装置、设备及存储介质
WO2020194596A1 (ja) * 2019-03-27 2020-10-01 日本食品製造合資会社 アイテム提案システム
CN110782276B (zh) * 2019-10-10 2022-04-26 微梦创科网络科技(中国)有限公司 一种访问分流策略干扰判断方法、装置及电子设备
CN112150195B (zh) * 2020-09-15 2024-07-05 珠海格力电器股份有限公司 一种目标物品的推荐方法、装置、设备及计算机可读介质
CN112203121A (zh) * 2020-10-12 2021-01-08 广州欢网科技有限责任公司 一种互联网电视广告投放方法和装置及系统
CN112231567A (zh) * 2020-10-20 2021-01-15 王明烨 一种精确推送信息的方法及装置
CN112364259A (zh) * 2020-11-24 2021-02-12 深圳市元征科技股份有限公司 一种信息推荐方法、装置、设备及介质
US20230325735A1 (en) * 2022-03-25 2023-10-12 Microsoft Technology Licensing, Llc Generating and processing contextualized group data

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101136938A (zh) * 2007-09-10 2008-03-05 北京易路联动技术有限公司 一种移动互联网应用的集中式管理方法和平台系统
CN102130933A (zh) * 2010-01-13 2011-07-20 中国移动通信集团公司 一种基于移动互联网的推荐方法、系统和设备
US20110179062A1 (en) * 2010-01-19 2011-07-21 Electronics And Telecommunications Research Institute Apparatus and method for sharing social media content

Family Cites Families (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7409362B2 (en) * 2004-12-23 2008-08-05 Diamond Review, Inc. Vendor-driven, social-network enabled review system and method with flexible syndication
US20090182589A1 (en) * 2007-11-05 2009-07-16 Kendall Timothy A Communicating Information in a Social Networking Website About Activities from Another Domain
US8073794B2 (en) * 2007-12-20 2011-12-06 Yahoo! Inc. Social behavior analysis and inferring social networks for a recommendation system
US8417698B2 (en) * 2008-05-06 2013-04-09 Yellowpages.Com Llc Systems and methods to provide search based on social graphs and affinity groups
US10269021B2 (en) * 2009-04-20 2019-04-23 4-Tell, Inc. More improvements in recommendation systems
US8671029B2 (en) * 2010-01-11 2014-03-11 Ebay Inc. Method, medium, and system for managing recommendations in an online marketplace
US20110202406A1 (en) * 2010-02-16 2011-08-18 Nokia Corporation Method and apparatus for distributing items using a social graph
US20110238608A1 (en) * 2010-03-25 2011-09-29 Nokia Corporation Method and apparatus for providing personalized information resource recommendation based on group behaviors
US20120166935A1 (en) * 2010-12-24 2012-06-28 Trademarkia, Inc. Automatic association of government brand information with domain and social media availability
CN102317941A (zh) * 2011-07-30 2012-01-11 华为技术有限公司 信息推荐方法、推荐引擎及网络系统
US20130054407A1 (en) * 2011-08-30 2013-02-28 Google Inc. System and Method for Recommending Items to Users Based on Social Graph Information

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101136938A (zh) * 2007-09-10 2008-03-05 北京易路联动技术有限公司 一种移动互联网应用的集中式管理方法和平台系统
CN102130933A (zh) * 2010-01-13 2011-07-20 中国移动通信集团公司 一种基于移动互联网的推荐方法、系统和设备
US20110179062A1 (en) * 2010-01-19 2011-07-21 Electronics And Telecommunications Research Institute Apparatus and method for sharing social media content

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110737846A (zh) * 2019-10-28 2020-01-31 北京字节跳动网络技术有限公司 一种社交界面推荐方法、装置、电子设备及存储介质
CN110737846B (zh) * 2019-10-28 2022-05-31 北京字节跳动网络技术有限公司 一种社交界面推荐方法、装置、电子设备及存储介质
US20220172260A1 (en) * 2020-02-07 2022-06-02 Tencent Technology (Shenzhen) Company Limited Method, apparatus, storage medium, and device for generating user profile
US12020267B2 (en) * 2020-02-07 2024-06-25 Tencent Technology (Shenzhen) Company Limited Method, apparatus, storage medium, and device for generating user profile
CN112950325A (zh) * 2021-03-16 2021-06-11 山西大学 一种社交行为融合的自注意力序列推荐方法
CN112950325B (zh) * 2021-03-16 2023-10-03 山西大学 一种社交行为融合的自注意力序列推荐方法
CN115348544A (zh) * 2021-05-14 2022-11-15 中国移动通信集团浙江有限公司 异网用户定位方法、终端设备及存储介质

Also Published As

Publication number Publication date
US20140330653A1 (en) 2014-11-06
CN103514204B (zh) 2018-11-20
US20210133817A1 (en) 2021-05-06
CN103514204A (zh) 2014-01-15

Similar Documents

Publication Publication Date Title
WO2014000431A1 (zh) 信息推荐方法和装置
US11632407B2 (en) Supplementing user web-browsing
US10007936B1 (en) Product review platform based on social connections
JP6408662B2 (ja) 自然言語処理に基づく、種々のオブジェクトに対する係数付与
US8909569B2 (en) System and method for revealing correlations between data streams
US8250096B2 (en) Access to trusted user-generated content using social networks
US10614143B2 (en) Systems and methods for automated page category recommendation
US10785181B2 (en) Sharing content to multiple public and private targets in a social network
US10305944B1 (en) Customizing content in a social stream
KR102138570B1 (ko) 가치 모델에 기반한 pymk의 사용과 컨텐츠를 결정하는 방법 및 시스템
US20140344035A1 (en) Managing content recommendations for customers
Carter et al. Internet-enabled collective intelligence as a precursor and predictor of consumer behaviour
Malathi et al. Analysis and visualization of social media networks
Zhang et al. The role of key online reviews in affecting online hotel booking: an empirical investigation
Dhekane et al. Talash: Friend Finding In Federated Social Networks.
US10440143B2 (en) Contextual trust based recommendation graph
Xu et al. Trust-based context-aware mobile social network service recommendation
US11436566B2 (en) Techniques for presenting a contextual contact list
KR20170095998A (ko) 콜 투 액션을 생성, 선택, 표시 및 실행하기 위한 시스템 및 방법
Kabbany et al. Developing E-market model using social networks
Pagare et al. Social recommender system by embedding social regularization
Sponder et al. Network Analysis and Social Network Mapping
Musiał A Method for Analysis of Node Position inthe Network of Internet Users

Legal Events

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

Ref document number: 13809843

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 13809843

Country of ref document: EP

Kind code of ref document: A1