WO2017092581A1 - 一种用户数据共享的方法和设备 - Google Patents

一种用户数据共享的方法和设备 Download PDF

Info

Publication number
WO2017092581A1
WO2017092581A1 PCT/CN2016/106579 CN2016106579W WO2017092581A1 WO 2017092581 A1 WO2017092581 A1 WO 2017092581A1 CN 2016106579 W CN2016106579 W CN 2016106579W WO 2017092581 A1 WO2017092581 A1 WO 2017092581A1
Authority
WO
WIPO (PCT)
Prior art keywords
user
information
query
relationship
user account
Prior art date
Application number
PCT/CN2016/106579
Other languages
English (en)
French (fr)
Inventor
王峰伟
何慧梅
吴东杏
何帝君
林瑞华
Original Assignee
阿里巴巴集团控股有限公司
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 阿里巴巴集团控股有限公司 filed Critical 阿里巴巴集团控股有限公司
Priority to AU2016364120A priority Critical patent/AU2016364120A1/en
Priority to JP2018528317A priority patent/JP6553816B2/ja
Priority to KR1020187018756A priority patent/KR102086936B1/ko
Priority to SG11201804350WA priority patent/SG11201804350WA/en
Priority to EP16869891.8A priority patent/EP3361704A4/en
Publication of WO2017092581A1 publication Critical patent/WO2017092581A1/zh
Priority to US15/991,784 priority patent/US10673979B2/en
Priority to PH12018501167A priority patent/PH12018501167A1/en
Priority to AU2019101565A priority patent/AU2019101565A4/en
Priority to AU2020202605A priority patent/AU2020202605A1/en

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
    • G06F16/907—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/535—Tracking the activity of the user
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/2866—Architectures; Arrangements
    • H04L67/30—Profiles
    • H04L67/306—User profiles
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/3331—Query processing
    • G06F16/3332—Query translation
    • G06F16/3334—Selection or weighting of terms from queries, including natural language queries
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L65/00—Network arrangements, protocols or services for supporting real-time applications in data packet communication
    • H04L65/40—Support for services or applications
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks

Definitions

  • the present application relates to the field of computer technologies, and in particular, to a method and device for user data sharing.
  • Different types of business platforms continuously accumulate data information of users in different aspects.
  • the business platform of some e-commerce platforms will continuously precipitate the user's shopping consumption information
  • other search engine business platforms will gradually precipitate users' Search information
  • other financial platform business platforms will precipitate the user's deposit management information.
  • the same user entity will have different accounts on the websites of different types of business platforms, which will precipitate different information. How to associate user information of the same user between different business platforms, open up user data between different business platforms, and realize user behavior information sharing is a very valuable and very difficult thing.
  • the embodiment of the present invention provides a method and a device for sharing user data, so as to open up user behavior information for the same user entity between different service platforms, and implement sharing of the user behavior information.
  • a method for user data sharing includes the following steps:
  • the at least one user account and its relationship information related to the query medium are searched from the media network, and the relationship information of the user account includes: a relationship between the user account and the query medium. Strength, and the strength of the relationship between the user account and other user accounts;
  • the edge weight is the relationship strength score
  • a network device including:
  • An obtaining module configured to obtain query media information
  • a query module configured to use the query medium information obtained by the obtaining module to search for at least one user account and relationship information related to the query medium from a media network, where the relationship information of the user account includes: The strength of the relationship between the user account and the query medium, and the strength of the relationship between the user account and other user accounts;
  • a creating module configured to construct a local media network by using the at least one user account and the relationship information found by the query module, where an edge in the local media network includes a relationship between a medium and a user account, and The relationship between the user account and other user accounts, and the edge weight is the relationship strength score;
  • a determining module configured to determine a trusted account from the at least one user account by using the local media network
  • the processing module is configured to obtain user data of the trusted account, and output the user data corresponding to the query media information to implement user data sharing.
  • the embodiment of the present application discloses a method and a device for sharing user data, and querying at least one user account and relationship information according to the obtained query medium information, constructing a local media network according to the queried information, and using the a local media network, determining a trusted account from the at least one user account, acquiring user data of the trusted account, and outputting the user data corresponding to the query media information to implement user data sharing, thereby
  • the user behavior information of the same user entity is opened between the service platforms to realize user behavior information sharing.
  • FIG. 1 is a schematic flowchart of a method for user data sharing according to Embodiment 1 of the present application;
  • FIG. 2 is a schematic structural diagram of a local medium network according to Embodiment 1 of the present application.
  • FIG. 3 is a schematic structural diagram of a network device according to Embodiment 2 of the present application.
  • the database of different business platforms will record some identity information of the user (such as user mobile phone number, ID card, bank card, etc.) and environmental information (such as IP, WIFI, MAC, IMEI, etc.). These identity information and environmental information can be Collectively referred to as media information. Since the media information left by the same user on different service platforms is the same, it can be used to achieve data access. The so-called data access is based on the same media information as described above, and the user behavior information for the same user precipitated in the database of different service platforms is obtained from the database of different service platforms, so as to realize data sharing of the user behavior information.
  • identity information of the user such as user mobile phone number, ID card, bank card, etc.
  • environmental information such as IP, WIFI, MAC, IMEI, etc.
  • Finding users through a single medium is less accurate, especially for some common media information, such as IP, MAC, and even mobile phones. These media information usually find many accounts, and these accounts are not necessarily belong to The same user's. Therefore, although multiple accounts are found by using the single media information, since the accounts are not belonging to the same user, the obtained user behavior information will be from different users, so that the acquired data has low accuracy and cannot be shared. User behavior information precipitated by the same user in a database of different business platforms.
  • the user behavior information may include, for example, the user's shopping information, the user's search information, and the user's deposit information.
  • the technical solution proposed by the embodiment of the present application is to use multiple media information to lock the same user.
  • a user is locked with a plurality of media information, it is also confirmed that the media information belongs to the user. Further, the media information belonging to the same user is used to match different accounts on different service platforms, thereby obtaining different accounts left by the user on different service platforms, and corresponding accounts are in the corresponding service platforms.
  • the user behavior information precipitated in the database obtains the user behavior information of the same user in different business platforms.
  • each service platform can share user behavior information precipitated by the same user in a database of different service platforms. For example, using the media information belonging to the same user u to match the corresponding account in the service platform a, the service platform b, and the service platform c, that is, searching for the account left by the user u in the service platform a, the service platform b, and the service platform c.
  • FIG. 1 is a schematic flowchart of a method for user data sharing according to Embodiment 1 of the present application, where the method specifically includes the following steps:
  • Step S101 Acquire query media information.
  • Media refers to people or things that make a relationship between two parties (people or things). Common media include accounts, ID cards, mobile phones, mailboxes, bank cards, device codes, and address-related information.
  • the query medium refers to a medium used to query an account of a target user in a media network.
  • these media are more likely to be associated with a target user's account, so The accuracy of querying target users is also high.
  • Step S102 Searching, by using the query media information, the at least one user account and relationship information related to the query medium from the media network.
  • the relationship information of the user account includes: a relationship strength between the user account and the query medium, and a relationship strength between the user account and other user accounts.
  • the media network is a network including an association relationship between user accounts, a relationship between a user account and a medium, and the media network includes relationship strengths, user accounts, and media between any user accounts. The strength of the relationship between.
  • step S101 if at least two query media information are obtained, the operation of this step is specifically: searching, by using the at least two query media information, the respective media related to the corresponding query media. User accounts with their relationship information.
  • the following processing options may be included, and after the user network is searched for the relationship information of the user account corresponding to the corresponding query medium, the method may further include:
  • step S103 Determining whether the at least two query media information belong to the same user account, and if yes, determining the same user account as a trusted account; otherwise, performing step S103.
  • Step S103 Construct a local media network by using the found at least one user account and its relationship information.
  • the local media network includes vertices and edges
  • the vertices in the local media network include media or user accounts
  • the edges in the local media network include a relationship between a medium and a user account, and the user account and The relationship between other user accounts
  • the edge weight is the relationship strength score.
  • FIG. 2 a schematic structural diagram of a local media network created according to a specific application scenario according to an embodiment of the present application, and a corresponding creation process is as follows:
  • media information 1, media information 2, media information 3, and media information 4 are acquired.
  • the corresponding accounts A and B are found according to the media information 1
  • the corresponding accounts A and C are found according to the media information 2
  • the corresponding accounts A and C are found according to the media information 3
  • the corresponding information is found according to the media information 4.
  • the accounts A and C corresponding to the media information 2 correspond to the media information 3.
  • Accounts A, C, accounts D, E corresponding to the media information 4 construct a local media network as shown in Figure 2, wherein media information 1, media information 2, media information 3, media information 4, and accounts A, account B, account C and account D are different vertices.
  • the line between the vertices in the figure is the edge, indicating the relationship between the corresponding vertices.
  • Step S104 Determine, by using the local media network, a trusted account from the at least one user account.
  • the processing solution of this step is to perform a hybrid sorting on the at least one user account by using the local media network created in step S103, and determine the top N user accounts as trusted accounts.
  • Step 1 the edge weight normalization, normalizing the edge weights of the edges in the local media network, that is, in the local media network, respectively, between each query media information and the corresponding user account And the original edge weight information between each user account performs the edge weight information normalization operation.
  • edge weight information normalization operation can be implemented as follows:
  • the edge weight normalization operation is performed on the original edge weight between each query medium information and the user account corresponding thereto and the original edge weight between the user accounts according to the change rate information.
  • f(x) is the edge weight value obtained after the edge weight information normalization operation
  • x in "ax" represents the original edge weight value
  • a is used as a variable in the logistic regression formula, which can be obtained by the following formula:
  • x0 represents the minimum value of the original edge weight
  • x99% represents the value of the 99% quantile of the edge weight.
  • Step 2 Iteratively calculates the edge weights, and iteratively calculates the normalized edge weights until convergence, that is, according to the edge weight information normalization operation obtained in step 1 between the query media information and the corresponding user account.
  • the edge weights and the edge weights between the user accounts are iteratively calculated until the iteration results converge, and the edge weights between the query media information and the corresponding user accounts, and the edge weights between the user accounts are obtained.
  • the local media network is defined as G
  • n represents the node data in G
  • the node can be a user account or a medium.
  • the weight of each edge is recalculated by the following formula 4 (the calculation formula of the edge weight information in the k+1th iteration), and the formula 4:
  • Rk(a,b) represents the edge weight information of the node a and the node b in the kth iteration
  • Rk+1(a,b) represents the node k and the node b in the k+1th iteration
  • the edge weight information, C represents the convergence speed control parameter or the attenuation factor
  • represents the sum of the edge weight information of the neighbor list of node a
  • represents the edge weight of the neighbor list of node b
  • the sum value of the information, i represents the i-th, j represents the jth
  • Ii(a) represents the neighbor list information of the node a
  • Ij(b) represents the neighbor list information of the node b.
  • media information and accounts corresponding to media information are called nodes, other nodes connected to nodes are called neighbors of the node, and all neighbors are called neighbor lists).
  • the purpose of the iterative calculation of the edge weight is to make the edge weight information between the obtained nodes tend to be stable, that is, to converge, instead of the unrestricted iteration; therefore, in this embodiment, After performing the iterative calculation operation to the preset number of times, it is judged whether the edge weight information between the nodes obtained at this time satisfies the convergence condition.
  • whether the edge weight information between the nodes obtained at this time satisfies the convergence condition can be determined as follows:
  • Step 3 Determine the comprehensive weight information, and calculate the comprehensive weight of each user account or the comprehensive weight of each query medium by using the weight of each edge after convergence.
  • W(a) is the media weight.
  • the weights of different query media are different.
  • the ID card is generally more reliable than the result of the mobile phone query, and therefore, the weight is higher.
  • the weights should also be different, usually the media weights are specified according to business experience and understanding. Therefore, different media weights are set depending on the type of medium.
  • u denotes a user account u in the local medium network
  • S denotes a media information set S composed of media information corresponding to the user account u
  • Score(u) denotes comprehensive weight information of the user account u
  • W(a)*R(a, u) represents the edge weight information between the user account u and the media information corresponding thereto;
  • W(a)*W(b)*R(a,b) represents the side weight information between the media information corresponding to the user account u.
  • Step 4 comprehensive weight sorting, according to the comprehensive weight of each user account obtained by step 3 or the comprehensive weight of each query medium, sorting each user account or each query medium to determine a trusted account or a trusted medium.
  • the at least one user account may be mixed and sorted by using the local media network, and the top N user accounts are determined to be trusted accounts, and N is a positive integer.
  • the number one user account is the most trusted account, and the credibility is the comprehensive weight score of the user account.
  • the medium associated with the most trusted account is a trusted medium.
  • the credibility is the overall weighted score of the medium.
  • the medium related to the trusted user can be directly determined as a trusted medium, and the medium belongs to the trusted account.
  • the medium is considered untrustworthy, ie the medium does not belong to a trusted account.
  • Step S105 Acquire user data of the trusted account, and output the user data corresponding to the query media information to implement user data sharing.
  • the operation mode of implementing user data sharing may be specifically as follows:
  • the user behavior information of the user corresponding to the trusted account recorded in the different database is obtained according to the at least two media information corresponding to the trusted account, so as to share the user behavior information.
  • the user corresponding to the user corresponding to the trusted account recorded in the different database is obtained according to at least two media information corresponding to the trusted account to different databases.
  • other media information that does not correspond to the trusted account may also be obtained in the local media network described above, and the media information is used as non-trusted media information.
  • non-trusted media information is not untrusted media information. Therefore, it is also necessary to judge these non-trusted media information to determine whether it is untrusted media information.
  • whether the non-trusted media information is untrusted media information is implemented by:
  • Step A Acquire a similarity between the non-trusted medium information and the media information corresponding to the trusted account.
  • the similarity corresponds to the same number of adjacent edges between the non-trusted media information and the trusted media information associated therewith.
  • the greater the number of identical neighbors between the non-trusted media information and the trusted media information associated therewith, indicating that the non-trusted media information corresponds to the trusted account The higher the similarity between the media information; conversely, the smaller the number, the lower the similarity.
  • a trusted medium it can be directly determined and belongs to a trusted account. If there is a non-trusted medium and the similarity (R) of the trusted medium is less than a certain threshold (such as 0.01), the medium is considered to be Untrustworthy, that is, the medium is not a trusted account.
  • a certain threshold such as 0.01
  • Step B comparing the similarity with a preset similarity threshold
  • Step C The non-trusted media information whose similarity is less than the preset similarity threshold is used as the untrusted media information.
  • the edge weight information of the ID card is higher;
  • the media information because its accuracy rate (confidence) of querying users is different, its side weight information should also be different. Therefore, after the user behavior information of the user corresponding to the trusted account recorded in the different database is obtained according to at least two media information corresponding to the trusted account, the user behavior information of the user corresponding to the trusted account recorded in the different database may be determined.
  • the strength of the relationship between the media information corresponding to the trusted account and the trusted account, and the relationship strength refers to the accuracy rate when the user is queried according to the media information; specifically, it can be implemented as follows:
  • Step A acquiring second integrated edge weight information of each media information corresponding to the trusted account, where each second integrated edge weight information is composed of the media information and an account corresponding thereto Comprehensive edge weight information between accounts.
  • the foregoing second integrated edge weight information may be determined according to the following manner:
  • K denotes an account set K composed of accounts corresponding to the media information m
  • Score(m) denotes second integrated edge weight information
  • Score(a)*R(a,m) represents the edge weight information between the media information m and its corresponding user
  • Score(a)*Score(b)*R(a,b) represents the edge weight information between the users corresponding to the media information m.
  • Step B Sort each of the second integrated edge weight information.
  • Step C Determine the relationship strength between the media information and the trusted account according to the ranking result.
  • the embodiment of the present application discloses a method and a device for sharing user data, and querying at least one user account and relationship information according to the obtained query medium information, constructing a local media network according to the queried information, and using the a local media network, determining a trusted account from the at least one user account, acquiring user data of the trusted account, and outputting the user data corresponding to the query media information to implement user data sharing, thereby
  • the user behavior information of the same user entity is opened between the service platforms to realize user behavior information sharing.
  • the second embodiment of the present invention further provides a network device, as shown in FIG. 3, specifically including:
  • the obtaining module 31 is configured to obtain query media information.
  • the querying module 32 is configured to use the query medium information acquired by the obtaining module 31 to search for at least one user account and relationship information related to the query medium from the media network, where the relationship information of the user account includes: The strength of the relationship between the user account and the query medium, and the user account and other users The strength of the relationship between accounts;
  • a creating module 33 configured to construct a local media network by using the at least one user account and the relationship information found by the query module 32, where an edge in the local media network includes a relationship between a medium and a user account, and The relationship between the user account and other user accounts, and the edge weight is a relationship strength score;
  • a determining module 34 configured to determine a trusted account from the at least one user account by using the local media network
  • the processing module 35 is configured to obtain user data of the trusted account, and output the user data corresponding to the query media information to implement user data sharing.
  • the determining module 34 is specifically configured to:
  • the at least one user account is mixed and ranked, and the top N user accounts are determined to be trusted accounts, and N is a positive integer.
  • the determining module 34 specifically includes:
  • a normalization unit for normalizing edge weights of respective edges in the local media network
  • An iterative unit configured to iteratively calculate the edge weight normalized by the normalization unit until convergence;
  • a calculating unit configured to calculate an integrated weight of each user account or an integrated weight of each query medium by using each edge weight after convergence of the iterative unit;
  • a sorting unit configured to sort the at least one user account or each query medium according to the comprehensive weight of the respective user accounts calculated by the calculating unit or the comprehensive weight of each query medium, and determine a trusted account or a trusted medium.
  • the obtaining module 31 is specifically configured to acquire at least two query media information.
  • the querying module 32 is specifically configured to use the at least two query media information to respectively find, from the media network, the user accounts associated with the corresponding query media with their relationship information.
  • the creating module 33 is specifically configured to:
  • the embodiment of the present application discloses a method and a device for sharing user data, and querying at least one user account and relationship information according to the obtained query medium information, constructing a local media network according to the queried information, and using the a local media network, determining a trusted account from the at least one user account, obtaining the trusted account
  • User data is outputted as user data corresponding to the query media information to implement user data sharing, thereby opening user behavior information for the same user entity between different service platforms to implement user behavior information sharing.
  • the embodiments of the present invention may be implemented by hardware, or may be implemented by means of software plus a necessary general hardware platform.
  • the technical solution of the embodiment of the present invention may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (which may be a CD-ROM, a USB flash drive, a mobile hard disk, etc.).
  • a number of instructions are included to cause a computer device (which may be a personal computer, a server, or a network side device, etc.) to perform the methods described in various implementation scenarios of embodiments of the present invention.
  • modules in the apparatus in the implementation scenario may be distributed in the apparatus for implementing the scenario according to the implementation scenario description, or may be correspondingly changed in one or more devices different from the implementation scenario.
  • the modules of the above implementation scenarios may be combined into one module, or may be further split into multiple sub-modules.
  • serial numbers of the embodiments of the present invention are merely for the description, and do not represent the advantages and disadvantages of the implementation scenarios.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Theoretical Computer Science (AREA)
  • General Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • Computer Hardware Design (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Computing Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Computational Linguistics (AREA)
  • Multimedia (AREA)
  • Library & Information Science (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Information Transfer Between Computers (AREA)
  • Storage Device Security (AREA)

Abstract

本申请实施例公开了一种用户数据共享的方法和设备，根据获取到的查询媒介信息，查询相关的至少一个用户帐户及其关系信息，根据查询到的信息构建局部媒介网络，并利用所述局部媒介网络，从所述至少一个用户账户中确定可信账户，获取所述可信账户的用户数据，并作为所述查询媒介信息对应的用户数据输出，以实现用户数据共享，从而，将不同业务平台之间的针对同一个用户实体的用户行为信息打通，以实现用户行为信息共享。

Description

一种用户数据共享的方法和设备
本申请要求2015年12月01日递交的申请号为201510866768.1、发明名称为“一种用户数据共享的方法和设备”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及计算机技术领域，特别涉及一种用户数据共享的方法和设备。
背景技术
随着互联网和信息化的发展，用户在日常业务的进行过程中，会将信息应用于不同的业务平台中，例如：网络购物、网络行为、金融服务、保险业务等诸多方面，而在这些业务实现的过程中，不同的业务平台也会相应的对用户的信息进行记录和累计，包括业务执行记录以及由此衍生的用户行为信息。
不同类型的业务平台分别的不断地积累了用户的在不同方面的数据信息，比如，一些电商平台的业务平台会不断沉淀用户的购物消费信息，另外一些搜索引擎的业务平台会不断沉淀用户的搜索信息，另外一些理财平台的业务平台会沉淀用户的存款理财信息。同一个用户实体在不同类型的业务平台的网站上会有不同的账户，沉淀了不同的信息。如何将不同业务平台之间的同一用户的用户信息进行关联，打通不同业务平台之间的用户数据，实现用户行为信息共享，是一件非常有价值，也非常困难的事情。
但是，由于用户识别，信息交互，分类统计等诸多难题的存在，现有技术中还没有一种方案能够将不同业务平台之间的针对同一个用户实体的用户行为信息打通，实现用户行为信息共享。
因此，现有技术中亟待找到一种能够将不同业务平台之间的针对同一个用户实体的用户行为信息打通，以实现用户行为信息共享的方案。
发明内容
本申请实施例提供了一种用户数据共享的方法和设备，以实现将不同业务平台之间的针对同一个用户实体的用户行为信息打通，实现共享所述用户行为信息。
为了达到上述目的，一种用户数据共享的方法，包括以下步骤：
获取查询媒介信息；
利用所述查询媒介信息，从媒介网络中查找得到与所述查询媒介相关的至少一个用户账户及其关系信息，所述用户账户的关系信息包括：所述用户账户与该查询媒介之间的关系强度，以及，所述用户账户与其他用户账户之间的关系强度；
利用查找到的所述至少一个用户账户及其关系信息，构建局部媒介网络，所述局部媒介网络中的边包括媒介和用户账户的关系，以及，所述用户账户与其他用户账户之间的关系，边权重为关系强度分值；
利用所述局部媒介网络，从所述至少一个用户账户中确定可信账户；
获取所述可信账户的用户数据，并作为所述查询媒介信息对应的用户数据输出，以实现用户数据共享。
一种网络设备，包括：
获取模块，用于获取查询媒介信息；
查询模块，用于利用所述获取模块所获取的查询媒介信息，从媒介网络中查找得到与所述查询媒介相关的至少一个用户账户及其关系信息，所述用户账户的关系信息包括：所述用户账户与该查询媒介之间的关系强度，以及，所述用户账户与其他用户账户之间的关系强度；
创建模块，用于利用所述查询模块所查找到的所述至少一个用户账户及其关系信息，构建局部媒介网络，所述局部媒介网络中的边包括媒介和用户账户的关系，以及，所述用户账户与其他用户账户之间的关系，边权重为关系强度分值；
确定模块，用于利用所述局部媒介网络，从所述至少一个用户账户中确定可信账户；
处理模块，用于获取所述可信账户的用户数据，并作为所述查询媒介信息对应的用户数据输出，以实现用户数据共享。
与现有技术相比，本申请实施例至少具有以下优点：
本申请实施例公开了一种用户数据共享的方法和设备，根据获取到的查询媒介信息，查询相关的至少一个用户帐户及其关系信息，根据查询到的信息构建局部媒介网络，并利用所述局部媒介网络，从所述至少一个用户账户中确定可信账户，获取所述可信账户的用户数据，并作为所述查询媒介信息对应的用户数据输出，以实现用户数据共享，从而，将不同业务平台之间的针对同一个用户实体的用户行为信息打通，以实现用户行为信息共享。
附图说明
图1是本申请实施例一提供的一种用户数据共享的方法的流程示意图；
图2是本申请实施例一提供的一种局部媒介网络的结构示意图；
图3是本申请实施例二提供的一种网络设备的结构示意图。
具体实施方式
下面将结合本申请的附图，对本申请中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本申请的部分实施例，而不是全部的实施例。基于本申请的实施例，本领域技术人员在没有做出创造性劳动的前提下所获得的所有其他实施例，都属于本申请保护的范围。
不同的业务平台的数据库都会记录用户的一些身份信息(如用户手机号码，身份证，银行卡等)和环境信息(如使用的IP，WIFI，MAC，IMEI等)，这些身份信息和环境信息可统称为媒介信息。由于同一个用户在不同的业务平台留下的媒介信息是相同的，因此可以将其用来实现数据的打通。所谓的数据的打通是基于上述相同的媒介信息，从不同的业务平台的数据库中，获取在不同业务平台的数据库沉淀的针对同一个用户的用户行为信息，以实现该用户行为信息的数据共享。
通过单一媒介去查找用户，这样的方式准确率比较低，尤其是对于一些公用的媒介信息，如IP，MAC甚至手机等，利用这些媒介信息通常会找到很多账户，而这些账户又不一定是属于同一个用户的。因此，虽然用该单一媒介信息查找到了多个账户，但由于这些账户不是属于同一用户的，所以，所获得用户行为信息将是来自于不同用户的，这样获取的数据准确率低，无法实现共享同一用户在不同业务平台的数据库中所沉淀的用户行为信息。
其中，用户行为信息可以包括例如用户的购物信息、用户的搜索信息以及用户的存款信息等。
而且对于很多业务场景来说，准确率是非常重要的，而且一般情况下，都会有丰富的媒介信息，因此，更好的方案是能够用多个媒介信息去更好的查找用户或者说锁定用户。
本申请实施例所提出的技术方案要实现的正是用多个媒介信息去锁定同一个用户。 而当用多个媒介信息锁定一个用户之后，同时也即确认了这些媒介信息是属于该用户的。进一步的，再用这些属于同一个用户的媒介信息在不同业务平台去匹配不同的账号，从而获取该用户在不同业务平台留下的不同的账号，以及在相应的账号在所对应的业务平台的数据库中所沉淀的用户行为信息，即获取了同一个用户在不同业务平台中的用户行为信息。
通过以上的处理，使各个业务平台可以共享同一用户在不同业务平台的数据库中所沉淀的用户行为信息。例如，利用属于同一个用户u的媒介信息在业务平台a、业务平台b和业务平台c中匹配相应的账号，即在业务平台a、业务平台b和业务平台c中查找用户u留下的账号，进而，获取所查找到的账号在业务平台a、业务平台b和业务平台c的数据库中所沉淀的用户行为信息，从而，获取到了用户u在业务平台a、业务平台b和业务平台c中的用户行为信息，在业务平台a、业务平台b和业务平台c之间实现了用户u的用户行为信息的共享。
如图1所示，为本申请实施例一提供的一种用户数据共享的方法的流程示意图，该方法具体包括以下步骤：
步骤S101、获取查询媒介信息。
媒介是指使双方(人或事物)发生关系的人或事物，常见的媒介包括账户、身份证、手机、邮箱、银行卡、设备码、地址相关的信息等。
而在本申请实施例中，查询媒介是指用以在媒介网络中查询目标用户的帐户的媒介，当前，可以确定这些媒介与目标用户的账户存在关联性的可能性较大，所以，用以查询目标用户的准确性也较高。在实际的应用中，可以选择已验证与用户相匹配，或者具有排他性的媒介来作为查询媒介。
需要进一步说明的是，为了提高查询的准确度，也为了降低干扰信息或识别度较低的信息作为查询媒介所导致的查询结果错误，本申请实施例所提出的技术方案中，优选的需要获取两个以上的查询媒介用以后续的处理流程，具体查询媒介数量可以根据实际场景需要进行设置，这样的变化并不会影响本申请的保护范围。
步骤S102、利用所述查询媒介信息，从媒介网络中查找得到与所述查询媒介相关的至少一个用户账户及其关系信息。
其中，所述用户账户的关系信息包括：所述用户账户与该查询媒介之间的关系强度，以及，所述用户账户与其他用户账户之间的关系强度。
需要说明的是，所述媒介网络是包括用户账户之间的关联关系、用户账户与媒介之间的关联关系的网络，所述媒介网络中包括任意用户账户之间的关系强度、用户账户和媒介之间的关系强度。
在步骤S101中，如果获取了至少两个查询媒介信息，则本步骤的操作具体为：分别利用所述至少两个查询媒介信息，从媒介网络中查找得到与对应的所述查询媒介相关的各个用户账户以其关系信息。
需要说明的是，在一种具体的应用场景中，可以包括以下的处理选择，从媒介网络中查找得到与对应的所述查询媒介相关的各个用户账户以其关系信息之后，进一步可以包括：
判断所述至少两个查询媒介信息是否属于同一个用户账户，若是，则将确定的同一个用户账户作为可信账户；否则，执行步骤S103。
步骤S103、利用查找到的所述至少一个用户账户及其关系信息，构建局部媒介网络。
在实际应用场景中，局部媒介网络的创建流程如下：
根据用户账户与媒介之间的共现关系，确定用户账户和媒介之间的关系强度，根据用户之间的关系，确定用户账户之间的关系强度，其中，所述用户之间的关系包括：用户之间的社交关系、资金关系、共设备关系、共媒介关系。
具体的，所述局部媒介网络包括顶点和边，所述局部媒介网络中的顶点包括媒介或用户账户，所述局部媒介网络中的边包括媒介和用户账户的关系，以及，所述用户账户与其他用户账户之间的关系，边权重为关系强度分值。
如图2所示，为本申请实施例所提出的根据一个具体的应用场景所创建的局部媒介网络的结构示意图，相应的创建过程如下：
首先，获取媒介信息1、媒介信息2、媒介信息3、媒介信息4。
然后，在媒介网络中，根据媒介信息1找对应的账户A、B，根据媒介信息2找对应的账户A、C，根据媒介信息3找对应的账户A、C，根据媒介信息4找对应的账户D、E。
最后，根据媒介信息1、媒介信息2、媒介信息3、媒介信息4、以及与该媒介信息1对应的账户A、B，与该媒介信息2对应的账户A、C，与该媒介信息3对应的账户A、C，与该媒介信息4对应的账户D、E构建出如附图2所示的局部媒介网络，其中，媒介信息1、媒介信息2、媒介信息3、媒介信息4，以及账户A、账户B、账户C和账户D作为不同的顶点，图中各顶点之间的连线即为边，示意相应顶点之间关系。
步骤S104、利用所述局部媒介网络，从所述至少一个用户账户中确定可信账户。
具体的，本步骤的处理方案为利用步骤S103所创建的局部媒介网络，对所述至少一个用户账户进行混合排序，将排在前N个用户账户确定为可信账户。
具体的，在实际应用场景中，上述的混合排序的实现流程如下：
步骤1、边权重归一化，对所述局部媒介网络中的各个边的边权重进行归一化，即在所述局部媒介网络中，分别对各查询媒介信息和与其对应的用户账户之间、以及各用户账户之间的原始边权重信息进行边权重信息归一操作。
需要具体说明的是，上述的边权重信息归一操作可以通过如下的方式实现：
首先，分别获取所述各查询媒介信息和与其对应的用户账户之间的原始边权重、各用户账户之间的原始边权重在第一预设次数内的变化率信息；
根据该变化率信息对各查询媒介信息和与其对应的用户账户之间的原始边权重、各用户账户之间的原始边权重进行边权重归一化操作。
在具体的应用场景下，可参考如下公式实现：
公式一(逻辑回归公式)：
Figure PCTCN2016106579-appb-000001
其中，f(x)为进行边权重信息归一操作后所获得边权重值，“ax”中的x表示原始的边权重值，a作为逻辑回归公式中的变量，可以通过以下公式求得：
公式二(求导公式)：
Figure PCTCN2016106579-appb-000002
其中，x0表示原始边权重的最小值；x99％表示边权重的99％分位数的值。
步骤2、迭代计算边权重，对归一化后的边权重进行迭代计算，直至收敛，即根据步骤1的边权重信息归一操作所获得的各查询媒介信息和与其对应的用户账户之间的边权重、以及各用户账户之间的边权重进行迭代计算操作，直至迭代结果收敛后，获得各查询媒介信息和与其对应的用户账户之间的边权重、以及各用户账户之间的边权重。
为了方便描述，定义数学符号如下：局部媒介网络定义为G，n表示G中的节点数据，节点可以是用户账户，也可以是媒介。
在具体的应用场景下，上述的迭代操作可参考如下公式实现：
公式三(初始状态下，也即第k次迭代中的边权重信息计算公式)：
Figure PCTCN2016106579-appb-000003
在迭代过程中，通过如下公式四重新计算每条边的权重(第k+1次迭代中的边权重信息计算公式)，公式四：
Figure PCTCN2016106579-appb-000004
其中，Rk(a,b)表示节点a和节点b的在第k次迭代中的边权重信息、Rk+1(a,b)表示节点a和节点b的在第k+1次迭代中的边权重信息、C表示收敛速度控制参数或者叫衰减因子、|I(a)|表示节点a的邻居列表的边权重信息的和值、|I(b)|表示节点b的邻居列表的边权重信息的和值、i表示第i个、j表示第j个、Ii(a)表示节点a的邻居列表信息、Ij(b)表示节点b的邻居列表信息。(在局部媒介网络中，媒介信息以及与媒介信息相对应的账户称为节点，与节点相连接的其他节点称为该节点的邻居，所有邻居称为邻居列表)。
需要说明的是，由于对边权重进行迭代计算的目的是：使求得的节点之间的边权重信息趋于稳定也即收敛，而非无限制的迭代下去；因此，在本实施例中，在进行到预设次数的迭代计算操作之后，要判断此时获得的节点之间的边权重信息是否满足收敛条件。
具体的，在本实施例中，可通过如下方式确定此时获得的节点之间的边权重信息是否满足收敛条件：
当进行第一预设次数的迭代计算操作后，分别获取所述各用户账户与与其相对应的各媒介信息之间新的边权重的第一变化值、以及与所述各用户账户之间的边权重的第二变化值；
判断所述第一变化值与第二变化值的和值是否小于预设的第一和值阀值；
若小于，则停止进行迭代计算操作；
若不小于，则继续进行迭代计算操作。
步骤3、确定综合权重信息，利用收敛后的各个边权重，计算各个用户账户的综合权重或各查询媒介的综合权重。
根据所述用户账户之间的边权重和用户账户与相应的媒介之间的边权重信息，确定 各用户账户的综合权重信息和各查询媒介信息的综合权重信息。
在具体的应用场景下，可参考如下公式实现：
公式五(媒介权重信息的计算公式)：
Figure PCTCN2016106579-appb-000005
其中，W(a)为媒介权重。通常，不同的查询媒介的权重是不一样的，比如作为查询媒介，身份证比手机查询的结果一般都更加可信，因此，权重更高。不同类型的查询媒介，由于其查询用户的准确率(置信度)是不一样的，权重也应该是不同的，通常该媒介权重是根据业务经验和理解指定的。因此，根据媒介的不同类型，设置不同的媒介权重。
公式六(各用户账户的综合权重信息计算公式)：
Figure PCTCN2016106579-appb-000006
其中，u表示该局部媒介网络中的用户账户u，S表示与用户账户u相对应的媒介信息所组成的媒介信息集合S，Score(u)表示用户账户u的综合权重信息；
W(a)*R(a,u)表示用户账户u与与其相对应的各媒介信息之间的边权重信息；
W(a)*W(b)*R(a,b)表示与用户账户u相对应的媒介信息之间的边权重信息。
步骤4、综合权重排序，根据步骤3所得到的各个用户账户的综合权重或各查询媒介的综合权重，对各用户账户或各个查询媒介进行排序，确定可信账户或可信媒介。
具体的，可以利用所述局部媒介网络，对所述至少一个用户账户进行混合排序，将排在前N个用户账户确定为可信账户，N为正整数。
N可以根据实际场景需要所设置的数值，例如N＝1。排名第一的用户账户为最可信账户，可信度为该用户账户的综合权重分值。最可信账户相关的媒介为可信媒介，同样，可信度为该媒介的综合权重分值。
对于可信媒介的判定，可以直接将可信用户相关的媒介判定为可信媒介，且该媒介属于可信账户。
如果存在非可信媒介，且该非可信媒介和可信媒介的相似度(R)小于某个阈值(如0.01)，则认为该媒介是不可信的，即该媒介不属于可信账户。
步骤S105、获取所述可信账户的用户数据，并作为所述查询媒介信息对应的用户数据输出，以实现用户数据共享。
在具体的应用场景中，实现用户数据共享的操作方式可以具体为：
根据所述可信账户所对应的至少两个媒介信息到不同数据库中去获取所述不同数据库中所记录的该可信账户所对应的用户的用户行为信息，以实现共享所述用户行为信息。
在本实施例中，作为一个优选步骤，在根据所述可信账户所对应的至少两个媒介信息到不同数据库中去获取所述不同数据库中所记录的该可信账户所对应的用户的用户行为信息后，还可以在上述的局部媒介网络中获取不与所述可信账户相对应的其他媒介信息，将这些媒介信息作为非可信媒介信息。需要说明的是，非可信媒介信息并非不可信媒介信息，因此，还需要对这些非可信媒介信息进行判断，以确定其是否是不可信媒介信息。
具体的，可通过如下方式实现判断所述非可信媒介信息是否是不可信媒介信息：
步骤A：获取所述非可信媒介信息与所述与所述可信账户相对应的媒介信息之间的相似度。
其中，所述相似度是与所述非可信媒介信息与与其相关的可信媒介信息之间所具有的相同的邻边的数量相对应的。
优选的，所述非可信媒介信息与与其相关的可信媒介信息之间所具有的相同的邻边的数量越多表示所述非可信媒介信息与所述与所述可信账户相对应的媒介信息之间的相似度越高；反之，该数量越少表示相似度越低。
在实际应用中，对于可信媒介，可以直接判定且属于可信账户，如果存在非可信媒介且和可信媒介的相似度(R)小于某个阈值(如0.01)，则认为该媒介是不可信的，即该媒介不属于可信账户。
步骤B：将所述相似度与预设的相似度阀值进行比较；
步骤C：将所述相似度小于预设的相似度阀值的非可信媒介信息作为不可信媒介信息。
另一方面，由于不同的媒介信息与账户之间的边权重信息是不一样的，比如：身份证比手机的查询用户结果要更加可信，因此身份证的边权重信息更高；另外不同类型的媒介信息，由于其查询用户的准确率(置信度)是不一样的，其边权重信息也应该是不同的。因此，在根据所述可信账户所对应的至少两个媒介信息到不同数据库中去获取所述不同数据库中所记录的该可信账户所对应的用户的用户行为信息后，还可以确定与所 述可信账户相对应的各媒介信息与所述可信账户的关系强度，该关系强度指根据媒介信息查询用户时的准确率；具体的，可通过如下方式实现：
步骤A：分别获取与所述可信账户相对应的各媒介信息的第二综合边权重信息；其中，所述各第二综合边权重信息为所述各媒介信息与与其相对应的账户所组成的账户之间的综合边权重信息。
具体的，上述的第二综合边权重信息可以根据以下方式来确定：
公式七(第二综合边权重信息计算公式)：
Figure PCTCN2016106579-appb-000007
其中，m表示该局部媒介网络中的媒介信息m，K表示与该媒介信息m相对应的账户所组成的账户集合K，Score(m)表示第二综合边权重信息；
Score(a)*R(a,m)表示媒介信息m与其相对应的各用户之间的边权重信息；
Score(a)*Score(b)*R(a,b)表示与媒介信息m相对应的用户之间的边权重信息。
步骤B：对所述各第二综合边权重信息进行排序。
步骤C：根据排序结果确定所述各媒介信息与所述可信账户的关系强度。
与现有技术相比，本申请实施例至少具有以下优点：
本申请实施例公开了一种用户数据共享的方法和设备，根据获取到的查询媒介信息，查询相关的至少一个用户帐户及其关系信息，根据查询到的信息构建局部媒介网络，并利用所述局部媒介网络，从所述至少一个用户账户中确定可信账户，获取所述可信账户的用户数据，并作为所述查询媒介信息对应的用户数据输出，以实现用户数据共享，从而，将不同业务平台之间的针对同一个用户实体的用户行为信息打通，以实现用户行为信息共享。
基于与上述方法同样的发明构思，本发明实施例二还提出了一种网络设备，如图3所示，具体包括：
获取模块31，用于获取查询媒介信息；
查询模块32，用于利用所述获取模块31所获取的查询媒介信息，从媒介网络中查找得到与所述查询媒介相关的至少一个用户账户及其关系信息，所述用户账户的关系信息包括：所述用户账户与该查询媒介之间的关系强度，以及，所述用户账户与其他用户 账户之间的关系强度；
创建模块33，用于利用所述查询模块32所查找到的所述至少一个用户账户及其关系信息，构建局部媒介网络，所述局部媒介网络中的边包括媒介和用户账户的关系，以及，所述用户账户与其他用户账户之间的关系，边权重为关系强度分值；
确定模块34，用于利用所述局部媒介网络，从所述至少一个用户账户中确定可信账户；
处理模块35，用于获取所述可信账户的用户数据，并作为所述查询媒介信息对应的用户数据输出，以实现用户数据共享。
在具体的应用场景中，所述确定模块34，具体用于：
利用所述局部媒介网络，对所述至少一个用户账户进行混合排序，将排在前N个用户账户确定为可信账户，N为正整数。
其中，确定模块34具体包括：
归一化单元，用于对所述局部媒介网络中的各个边的边权重进行归一化；
迭代单元，用于对所述归一化单元归一化后的边权重进行迭代计算，直至收敛；
计算单元，用于利用所述迭代单元收敛后的各个边权重，计算各个用户账户的综合权重或各查询媒介的综合权重；
排序单元，用于根据所述计算单元计算的所述各个用户账户的综合权重或各查询媒介的综合权重，对所述至少一个用户账户或各个查询媒介排序，确定可信账户或可信媒介。
优选的，所述获取模块31，具体用于获取至少两个查询媒介信息；
所述查询模块32，具体用于分别利用所述至少两个查询媒介信息，从媒介网络中查找得到与对应的所述查询媒介相关的各个用户账户以其关系信息。
相应的，所述创建模块33，具体用于：
判断所述至少两个查询媒介信息是否属于同一个用户账户，若是，则将确定的同一个用户账户作为可信账户；否则，利用查找到的所述至少一个用户账户及其关系信息，构建局部媒介网络。
与现有技术相比，本申请实施例至少具有以下优点：
本申请实施例公开了一种用户数据共享的方法和设备，根据获取到的查询媒介信息，查询相关的至少一个用户帐户及其关系信息，根据查询到的信息构建局部媒介网络，并利用所述局部媒介网络，从所述至少一个用户账户中确定可信账户，获取所述可信账户 的用户数据，并作为所述查询媒介信息对应的用户数据输出，以实现用户数据共享，从而，将不同业务平台之间的针对同一个用户实体的用户行为信息打通，以实现用户行为信息共享。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到本发明实施例可以通过硬件实现，也可以借助软件加必要的通用硬件平台的方式来实现。基于这样的理解，本发明实施例的技术方案可以以软件产品的形式体现出来，该软件产品可以存储在一个非易失性存储介质(可以是CD-ROM，U盘，移动硬盘等)中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或网络侧设备等)执行本发明实施例各个实施场景所述的方法。
本领域技术人员可以理解附图只是一个优选实施场景的示意图，附图中的模块或流程并不一定是实施本发明实施例所必须的。
本领域技术人员可以理解实施场景中的装置中的模块可以按照实施场景描述进行分布于实施场景的装置中，也可以进行相应变化位于不同于本实施场景的一个或多个装置中。上述实施场景的模块可以合并为一个模块，也可以进一步拆分成多个子模块。
上述本发明实施例序号仅仅为了描述，不代表实施场景的优劣。
以上公开的仅为本发明实施例的几个具体实施场景，但是，本发明实施例并非局限于此，任何本领域的技术人员能思之的变化都应落入本发明实施例的业务限制范围。

Claims (14)

  1. 一种用户数据共享的方法，其特征在于，所述方法包括：
    获取查询媒介信息；
    利用所述查询媒介信息，从媒介网络中查找得到与所述查询媒介相关的至少一个用户账户及其关系信息，所述用户账户的关系信息包括：所述用户账户与该查询媒介之间的关系强度，以及，所述用户账户与其他用户账户之间的关系强度；
    利用查找到的所述至少一个用户账户及其关系信息，构建局部媒介网络，所述局部媒介网络中的边包括媒介和用户账户的关系，以及，所述用户账户与其他用户账户之间的关系，边权重为关系强度分值；
    利用所述局部媒介网络，从所述至少一个用户账户中确定可信账户；
    获取所述可信账户的用户数据，并作为所述查询媒介信息对应的用户数据输出，以实现用户数据共享。
  2. 如权利要求1所述的方法，其特征在于，所述利用所述局部媒介网络，从所述至少一个用户账户中确定可信账户，具体包括：
    利用所述局部媒介网络，对所述至少一个用户账户进行混合排序，将排在前N个用户账户确定为可信账户，N为正整数。
  3. 如权利要求2所述的方法，其特征在于，利用所述局部媒介网络，对所述至少一个用户账户进行混合排序，具体包括：
    对所述局部媒介网络中的各个边的边权重进行归一化；
    对归一化后的边权重进行迭代计算，直至收敛；
    利用收敛后的各个边权重，计算各个用户账户的综合权重或各查询媒介的综合权重；
    根据所述各个用户账户的综合权重或各查询媒介的综合权重，对所述至少一个用户账户或各个查询媒介进行排序，确定可信账户或可信媒介。
  4. 如权利要求3所述的方法，其特征在于，所述对所述局部媒介网络中的各个边的边权重进行归一化，具体包括：
    分别获取所述各查询媒介信息与与其对应的用户账户之间的原始边权重、各用户账户之间的原始边权重在第一预设次数内的变化率；
    根据所述变化率对所述各查询媒介信息与与其对应的用户账户之间的原始边权重、各用户账户之间的原始边权重进行边权重归一化操作。
  5. 如权利要求3所述的方法，其特征在于，所述对归一化后的边权重进行迭代计算，直至收敛，具体包括：
    根据所述边权重归一化操作之后所获得的各查询媒介信息与与其相对应的用户账户之间的边权重、各用户账户之间的边权重进行迭代计算操作，以分别确定所述各用户账户与与其对应的查询媒介信息之间的边权重、以及所述各用户账户之间的边权重；
    当进行第一预设次数的迭代计算操作后，分别获取所述各用户账户与与其相对应的各查询媒介信息之间新的边权重的第一变化值、以及与所述各用户账户之间的边权重的第二变化值；
    判断所述第一变化值与第二变化值的和值是否小于预设的和值阀值；
    若小于，则停止进行迭代计算操作；
    若不小于，则继续进行迭代计算操作。
  6. 如权利要求1所述的方法，其特征在于，所述获取查询媒介信息，具体包括：
    获取至少两个查询媒介信息；
    利用所述查询媒介信息，从媒介网络中查找得到与所述查询媒介相关的至少一个用户账户及其关系信息，具体包括：
    分别利用所述至少两个查询媒介信息，从媒介网络中查找得到与对应的所述查询媒介相关的各个用户账户以其关系信息。
  7. 如权利要求6所述的方法，其特征在于，在分别利用所述至少两个查询媒介信息，从媒介网络中查找得到与对应的所述查询媒介相关的各个用户账户以其关系信息之后，还包括：
    判断所述至少两个查询媒介信息是否属于同一个用户账户，若是，则将确定的同一个用户账户作为可信账户；否则，利用查找到的所述至少一个用户账户及其关系信息，构建局部媒介网络。
  8. 如权利要求1所述的方法，其特征在于，所述媒介网络是包括用户账户之间的关联关系、用户账户与媒介之间的关联关系的网络，所述媒介网络中包括任意用户账户之间的关系强度、用户账户和媒介之间的关系强度。
  9. 如权利要求8所述的方法，其特征在于，所述媒介网络的建立方法包括：
    根据用户账户与媒介之间的共现关系，确定用户账户和媒介之间的关系强度；
    根据用户之间的关系，确定用户账户之间的关系强度，其中所述用户之间的关系包括：用户之间的社交关系、资金关系、共设备关系、共媒介关系。
  10. 一种网络设备，其特征在于，包括：
    获取模块，用于获取查询媒介信息；
    查询模块，用于利用所述获取模块所获取的查询媒介信息，从媒介网络中查找得到与所述查询媒介相关的至少一个用户账户及其关系信息，所述用户账户的关系信息包括：所述用户账户与该查询媒介之间的关系强度，以及，所述用户账户与其他用户账户之间的关系强度；
    创建模块，用于利用所述查询模块所查找到的所述至少一个用户账户及其关系信息，构建局部媒介网络，所述局部媒介网络中的边包括媒介和用户账户的关系，以及，所述用户账户与其他用户账户之间的关系，边权重为关系强度分值；
    确定模块，用于利用所述局部媒介网络，从所述至少一个用户账户中确定可信账户；
    处理模块，用于获取所述可信账户的用户数据，并作为所述查询媒介信息对应的用户数据输出，以实现用户数据共享。
  11. 如权利要求10所述的设备，其特征在于，所述确定模块，具体用于：
    利用所述局部媒介网络，对所述至少一个用户账户进行混合排序，将排在前N个用户账户确定为可信账户，N为正整数。
  12. 如权利要求11所述的设备，其特征在于，所述确定模块具体包括：
    归一化单元，用于对所述局部媒介网络中的各个边的边权重进行归一化；
    迭代单元，用于对所述归一化单元归一化后的边权重进行迭代计算，直至收敛；
    计算单元，用于利用所述迭代单元收敛后的各个边权重，计算各个用户账户的综合权重或各查询媒介的综合权重；
    排序单元，用于根据所述计算单元计算的所述各个用户账户的综合权重或各查询媒介的综合权重，对所述至少一个用户账户或各个查询媒介排序，确定可信账户或可信媒介。
  13. 如权利要求10所述的设备，其特征在于，
    所述获取模块，具体用于获取至少两个查询媒介信息；
    所述查询模块，具体用于分别利用所述至少两个查询媒介信息，从媒介网络中查找得到与对应的所述查询媒介相关的各个用户账户以其关系信息。
  14. 如权利要求13所述的设备，其特征在于，所述创建模块，具体用于：
    判断所述至少两个查询媒介信息是否属于同一个用户账户，若是，则将确定的同一 个用户账户作为可信账户；否则，利用查找到的所述至少一个用户账户及其关系信息，构建局部媒介网络。
PCT/CN2016/106579 2015-12-01 2016-11-21 一种用户数据共享的方法和设备 WO2017092581A1 (zh)

Priority Applications (9)

Application Number Priority Date Filing Date Title
AU2016364120A AU2016364120A1 (en) 2015-12-01 2016-11-21 User data sharing method and device
JP2018528317A JP6553816B2 (ja) 2015-12-01 2016-11-21 ユーザデータ共有方法及び装置
KR1020187018756A KR102086936B1 (ko) 2015-12-01 2016-11-21 사용자 데이터 공유 방법 및 디바이스
SG11201804350WA SG11201804350WA (en) 2015-12-01 2016-11-21 User data sharing method and device
EP16869891.8A EP3361704A4 (en) 2015-12-01 2016-11-21 METHOD AND DEVICE FOR SHARING USER DATA
US15/991,784 US10673979B2 (en) 2015-12-01 2018-05-29 User data sharing method and device
PH12018501167A PH12018501167A1 (en) 2015-12-01 2018-06-01 User data sharing method and device
AU2019101565A AU2019101565A4 (en) 2015-12-01 2019-12-12 User data sharing method and device
AU2020202605A AU2020202605A1 (en) 2015-12-01 2020-04-17 User data sharing method and device

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510866768.1A CN106817390B (zh) 2015-12-01 2015-12-01 一种用户数据共享的方法和设备
CN201510866768.1 2015-12-01

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US15/991,784 Continuation US10673979B2 (en) 2015-12-01 2018-05-29 User data sharing method and device

Publications (1)

Publication Number Publication Date
WO2017092581A1 true WO2017092581A1 (zh) 2017-06-08

Family

ID=58796341

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/106579 WO2017092581A1 (zh) 2015-12-01 2016-11-21 一种用户数据共享的方法和设备

Country Status (10)

Country Link
US (1) US10673979B2 (zh)
EP (1) EP3361704A4 (zh)
JP (1) JP6553816B2 (zh)
KR (1) KR102086936B1 (zh)
CN (1) CN106817390B (zh)
AU (3) AU2016364120A1 (zh)
MY (1) MY178354A (zh)
PH (1) PH12018501167A1 (zh)
SG (1) SG11201804350WA (zh)
WO (1) WO2017092581A1 (zh)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107766322A (zh) * 2017-08-31 2018-03-06 平安科技（深圳）有限公司 同名实体识别方法、电子设备及计算机可读存储介质
CN108510312B (zh) * 2018-03-07 2021-06-29 创新先进技术有限公司 多业务共享召回模式下的业务召回方法、系统和服务器
CN110248217B (zh) * 2019-07-08 2022-04-22 中国联合网络通信集团有限公司 一种用户数据同步方法及装置
CN111612548B (zh) * 2020-05-27 2023-08-29 恩亿科(北京)数据科技有限公司 信息获取方法、装置、计算机设备和可读存储介质
CN115225635A (zh) * 2022-06-30 2022-10-21 浪潮云信息技术股份公司 一种多云纳管方法及装置

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102209335A (zh) * 2011-05-24 2011-10-05 四川奥诚科技有限责任公司 一种多网元用户数据的获取方法及装置
US20140059658A1 (en) * 2012-08-22 2014-02-27 Mcafee, Inc. Privacy broker
CN103780690A (zh) * 2014-01-20 2014-05-07 北京集奥聚合科技有限公司 用户数据的安全共享方法和系统
CN104978383A (zh) * 2015-02-12 2015-10-14 腾讯科技（深圳）有限公司 一种数据互通的方法,及数据互通设备

Family Cites Families (33)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7225156B2 (en) 2001-07-11 2007-05-29 Fisher Douglas C Persistent dynamic payment service
PL374305A1 (en) 2001-12-28 2005-10-03 Jeffrey James Jonas Real time data warehousing
US7221935B2 (en) 2002-02-28 2007-05-22 Telefonaktiebolaget Lm Ericsson (Publ) System, method and apparatus for federated single sign-on services
US8597116B2 (en) 2002-03-12 2013-12-03 Igt Virtual player tracking and related services
US20030195857A1 (en) 2002-04-10 2003-10-16 Alessandro Acquisti Communication technique to verify and send information anonymously among many parties
US20050108575A1 (en) 2003-11-18 2005-05-19 Yung Chong M. Apparatus, system, and method for faciliating authenticated communication between authentication realms
US7519822B2 (en) 2004-03-10 2009-04-14 Hewlett-Packard Development Company, L.P. Method and apparatus for processing descriptive statements
US7562382B2 (en) 2004-12-16 2009-07-14 International Business Machines Corporation Specializing support for a federation relationship
US8184811B1 (en) 2005-10-12 2012-05-22 Sprint Spectrum L.P. Mobile telephony content protection
US20070226152A1 (en) 2006-03-21 2007-09-27 Austin Jones System and method for anonymous transactions and conveyances
US8108266B2 (en) 2007-01-29 2012-01-31 Hewlett-Packard Development Company, L.P. Methods for providing secure eCommerce transactions
US8560630B2 (en) * 2007-02-28 2013-10-15 Microsoft Corporation Sharing data over trusted networks
CN102349276B (zh) 2009-03-12 2014-07-02 Nec欧洲有限公司 支持用户或实体的分布式数据的管理和交换的方法
US8392266B2 (en) 2009-11-13 2013-03-05 Omnione Usa, Inc. System and method for certifying information relating to transactions between a seller and a purchaser
US8825962B1 (en) * 2010-04-20 2014-09-02 Facebook, Inc. Push-based cache invalidation notification
US8918848B2 (en) 2010-04-26 2014-12-23 Blackberry Limited Method and system for third party client authentication
US8402527B2 (en) 2010-06-17 2013-03-19 Vmware, Inc. Identity broker configured to authenticate users to host services
US8762284B2 (en) 2010-12-16 2014-06-24 Democracyontheweb, Llc Systems and methods for facilitating secure transactions
AU2012223415B2 (en) 2011-02-28 2017-05-18 Visa International Service Association Secure anonymous transaction apparatuses, methods and systems
US8965848B2 (en) * 2011-08-24 2015-02-24 International Business Machines Corporation Entity resolution based on relationships to a common entity
US20130124327A1 (en) * 2011-11-11 2013-05-16 Jumptap, Inc. Identifying a same user of multiple communication devices based on web page visits
US9444726B2 (en) * 2012-07-31 2016-09-13 At&T Intellectual Property I, L.P. Distributing communication of a data stream among multiple devices
US10332108B2 (en) 2012-08-01 2019-06-25 Visa International Service Association Systems and methods to protect user privacy
CN102821109B (zh) * 2012-08-28 2015-06-03 腾讯科技（深圳）有限公司 在即时通信应用中实现数据共享的方法、相关设备及系统
US9098819B1 (en) 2012-10-18 2015-08-04 Google Inc. Identifying social network accounts belonging to the same user
US20140129459A1 (en) * 2012-11-06 2014-05-08 Niklas Baudy Method and system for social media integration into business applications
US10559013B2 (en) * 2013-03-07 2020-02-11 Facebook, Inc. Identifying users for advertising opportunities based on paired identifiers
KR102205531B1 (ko) * 2013-12-04 2021-01-20 네이버 주식회사 소셜 네트워크 서비스의 사용자 관계정보에 기반한 지식 공유 서비스 제공 시스템 및 방법
GB2524073A (en) * 2014-03-14 2015-09-16 Ibm Communication method and system for accessing media data
CN105024984B (zh) * 2014-04-30 2017-12-01 腾讯科技（深圳）有限公司 权限设置方法、装置及系统
CN105516068B (zh) * 2014-09-26 2019-06-04 阿里巴巴集团控股有限公司 信息交互以及人体生物特征数据与账号的关联方法、装置
TWI564729B (zh) * 2015-08-07 2017-01-01 廣達電腦股份有限公司 資料分享系統以及其方法
CN105430487A (zh) * 2015-11-30 2016-03-23 小米科技有限责任公司 一种播放多媒体数据的方法、装置及系统

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102209335A (zh) * 2011-05-24 2011-10-05 四川奥诚科技有限责任公司 一种多网元用户数据的获取方法及装置
US20140059658A1 (en) * 2012-08-22 2014-02-27 Mcafee, Inc. Privacy broker
CN103780690A (zh) * 2014-01-20 2014-05-07 北京集奥聚合科技有限公司 用户数据的安全共享方法和系统
CN104978383A (zh) * 2015-02-12 2015-10-14 腾讯科技（深圳）有限公司 一种数据互通的方法,及数据互通设备

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3361704A4 *

Also Published As

Publication number Publication date
KR102086936B1 (ko) 2020-03-09
PH12018501167A1 (en) 2019-01-21
EP3361704A1 (en) 2018-08-15
AU2020202605A1 (en) 2020-05-14
AU2016364120A1 (en) 2018-06-07
US20180278716A1 (en) 2018-09-27
JP2019504393A (ja) 2019-02-14
CN106817390A (zh) 2017-06-09
MY178354A (en) 2020-10-08
US10673979B2 (en) 2020-06-02
SG11201804350WA (en) 2018-06-28
JP6553816B2 (ja) 2019-07-31
KR20180089479A (ko) 2018-08-08
CN106817390B (zh) 2020-04-24
EP3361704A4 (en) 2019-06-19
AU2019101565A4 (en) 2020-01-23

Similar Documents

Publication Publication Date Title
WO2017092581A1 (zh) 一种用户数据共享的方法和设备
WO2022126971A1 (zh) 基于密度的文本聚类方法、装置、设备及存储介质
US20200272645A1 (en) Identity resolution in big, noisy, and/or unstructured data
US11727053B2 (en) Entity recognition from an image
US9116879B2 (en) Dynamic rule reordering for message classification
US20130218620A1 (en) Method and system for skill extraction, analysis and recommendation in competency management
CN111046237B (zh) 用户行为数据处理方法、装置、电子设备及可读介质
WO2013097026A1 (en) Systems and methods for visualizing social graphs
CN104077723B (zh) 一种社交网络推荐系统及方法
CN109117669B (zh) MapReduce相似连接查询的隐私保护方法及系统
WO2020181908A1 (zh) 风险决策方法及装置
US11985153B2 (en) System and method for detecting anomalous activity based on a data distribution
CN115238062A (zh) 技术产权匹配方法和系统
US20220019934A1 (en) System for artificial intelligence-based electronic data analysis in a distributed server network
TWI666559B (zh) 用戶資料共用的方法和設備
US11921787B2 (en) Identity-aware data management
US11556336B2 (en) System for computer code development environment cloning and authentication using a distributed server network and machine learning
US11296882B1 (en) System for intelligent identification of unauthorized users in a distributed register network
US20240195913A1 (en) System and method for classifying calls
CN114218447A (zh) 一种图处理方法和系统

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: 16869891

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 2016869891

Country of ref document: EP

WWE Wipo information: entry into national phase

Ref document number: 11201804350W

Country of ref document: SG

WWE Wipo information: entry into national phase

Ref document number: 2018528317

Country of ref document: JP

WWE Wipo information: entry into national phase

Ref document number: 12018501167

Country of ref document: PH

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2016364120

Country of ref document: AU

Date of ref document: 20161121

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 20187018756

Country of ref document: KR

Kind code of ref document: A