WO2019080403A1 - 社交平台用户的现实关系匹配方法、装置及可读存储介质 - Google Patents

社交平台用户的现实关系匹配方法、装置及可读存储介质

Info

Publication number
WO2019080403A1
WO2019080403A1 PCT/CN2018/075204 CN2018075204W WO2019080403A1 WO 2019080403 A1 WO2019080403 A1 WO 2019080403A1 CN 2018075204 W CN2018075204 W CN 2018075204W WO 2019080403 A1 WO2019080403 A1 WO 2019080403A1
Authority
WO
WIPO (PCT)
Prior art keywords
node
relationship
user
seed user
closeness
Prior art date
Application number
PCT/CN2018/075204
Other languages
English (en)
French (fr)
Inventor
王健宗
黄章成
吴天博
肖京
Original Assignee
平安科技(深圳)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 平安科技(深圳)有限公司 filed Critical 平安科技(深圳)有限公司
Publication of WO2019080403A1 publication Critical patent/WO2019080403A1/zh

Links

Images

Classifications

    • 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
    • G06F16/958Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking
    • 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
    • G06Q50/00Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/01Social networking

Definitions

  • the present application relates to the field of computer network technologies, and in particular, to a real-world relationship matching method for a social platform user, a data processing device, and a computer-readable storage medium.
  • the usual method is to measure the distance between nodes according to the network structure, or use the clustering method to find the cluster to divide the nodes, and calculate the similarity between users through different algorithms in the social topology network structure. To determine the relationship between users.
  • the clustering method to find the cluster to divide the nodes, and calculate the similarity between users through different algorithms in the social topology network structure. To determine the relationship between users.
  • due to the characteristics of social networks although many nodes are close to each other, they may be just online friends, and they have never met in offline or real life.
  • the main purpose of the present application is to provide a real-life relationship matching method, a data processing device, and a computer-readable storage medium for an online social user, aiming at solving a technique for accurately calculating and identifying an offline user relationship of an online user in real life. problem.
  • a real-life relationship matching method for a social platform user includes the following steps:
  • Providing a target user of a social platform defining the target user as a seed user node in a three-dimensional space model, defining other users on the social platform as neighbor nodes of the seed user, and defining a pseudo around the seed user a node; wherein the pseudo node is defined to be in contact with the seed user node and is defined as not associated with other online friends of the seed user on the social platform;
  • Determining, by the neighboring node of the seed user node that is calculated by the random walk algorithm, a neighboring node that is closer to the seed user node than the seed user node and the pseudo node, and the target node The user is an offline friend relationship.
  • the application also provides a data processing device, including:
  • a model node definition module which provides a target user of a social platform, defines the target user as a seed user node in a three-dimensional space model, and defines other users on the social platform as neighbor nodes of the seed user, in the seed
  • a pseudo node is defined around the user; wherein the pseudo node and the seed user are defined to be in contact with each other and are not associated with other online friends of the seed user;
  • the tightness calculation module calculates, by a preset random walk algorithm, a closeness between the seed user node and a neighbor node corresponding to other users on the social platform, and a closeness between the seed user node and the pseudo node;
  • the tightness judging module determines whether the closeness of the neighboring node of the seed user node calculated by the random walk algorithm and the seed user node is greater than the closeness of the seed user node and the pseudo node;
  • the offline relationship judging module the neighboring node of the seed user node calculated by the random walk algorithm is closer to the seed user node than the neighbor node of the seed user node and the pseudo node Determined to be an offline friend relationship with the target user.
  • the present application also provides a data processing apparatus comprising a memory, a processor, and a computer program stored on the memory and operable on the processor, the processor executing the program to implement the social platform of any of the above The steps of the user's realistic relationship matching method.
  • the application further provides a computer readable storage medium having stored thereon a computer program, the program being executed by the processor to implement the steps of the social network user's real relationship matching method according to any of the above.
  • the target user is defined as a seed user node in a three-dimensional space model, and other users on the social platform are defined as neighbor nodes of the seed user, Defining a pseudo node around the seed user; wherein the pseudo node is defined to be in mutual interest with the seed user node, and is defined as not associated with other online friends of the seed user on the social platform; Setting a random walk algorithm to calculate a closeness between the seed user node and a neighbor node corresponding to other users on the social platform, and a closeness between the seed user node and the pseudo node; determining the random walk algorithm Calculating whether the closeness of the neighboring node and the seed user node is greater than the closeness of the seed user node and the pseudo node; and the neighboring node of the seed user node calculated by the random walk algorithm
  • the tightness with the seed user node is greater than the tightness of the seed user node and the pseudo node
  • the neighbor node is determined to be an
  • FIG. 1 is a flowchart of a method for matching a real relationship of a social platform user in a first embodiment of the present application
  • FIG. 2 is a flowchart of a method for matching a real relationship of a social platform user in a second embodiment of the present application
  • FIG. 3 is a flowchart of a method for matching a real relationship of a social platform user in a third embodiment of the present application
  • FIG. 4 is a flowchart of a method for matching a real relationship of a social platform user in a fourth embodiment of the present application
  • FIG. 5 is a schematic structural diagram of hardware modules of a data processing apparatus according to an embodiment of the present application.
  • FIG. 6 is a schematic structural diagram of functional modules of a data processing apparatus according to an embodiment of the present application.
  • FIG. 1 is a flowchart of a method for a real-world relationship matching method 100 of a social platform user in a first embodiment of the present application, where the real-life relationship matching method 100 of the social platform user includes the following steps:
  • Step S10 providing a target user of a social platform, defining the target user as a seed user node in a three-dimensional space model, and defining other users on the social platform as neighbor nodes of the seed user, around the seed user
  • a pseudo node is defined; wherein the pseudo node is defined to be in mutual interest with the seed user node and is defined as not associated with other online friends of the seed user on the social platform.
  • the target users on the social platform may be registered users on the same platform, or may be users of different social platforms on the social platform.
  • the social platform may be various social softwares or social networking websites in the Internet, and is not specifically limited herein.
  • Step S20 Calculate the closeness between the seed user node and the neighbor node corresponding to other users on the social platform by using a preset random walk algorithm, and the closeness of the seed user node and the pseudo node.
  • the random walk algorithm is an algorithm commonly used to calculate the probability of randomly moving from one node to another in a one-dimensional/two-dimensional/three-dimensional network, and thus can be used to measure between nodes.
  • the tightness is an algorithm commonly used to calculate the probability of randomly moving from one node to another in a one-dimensional/two-dimensional/three-dimensional network, and thus can be used to measure between nodes. The tightness.
  • the pseudo node is an imaginary imaginary node that is in contact with the seed user, and is equivalent to a preset value.
  • the pseudo node is out to distinguish the offline user
  • the pseudo node is defined as mutual attention with the seed user node, and is defined as other with the seed user on the social platform.
  • Online friends are not related; the pseudo-nodes here simulate an online friend who is concerned with the seed user by coincidence. Therefore, the closeness of the seed user and the pseudo-node can be used as a threshold to distinguish the offline friends.
  • the probability of walking from one node to another in the three-dimensional space model may be random or non-random with a certain weight.
  • the prob function is a function of the corresponding probability that the value falls within the specified interval, i ⁇ (1, n).
  • Step S30 determining whether the closeness of the neighboring node and the seed user node calculated by the random walk algorithm is greater than the closeness of the seed user node and the pseudo node;
  • Step S40 determining, by using the random walk algorithm, the neighbor node of the seed user node that is closer to the seed user node than the neighbor node of the seed user node and the pseudo node is determined to be
  • the target user is an offline friend relationship.
  • the pseudo node is a hypothetical node for distinguishing between offline friends and non-offline friends, and the pseudo node only pays attention to the target user, so all neighbor nodes calculated based on the random walk algorithm and the A user whose seed user node is more tight than the pseudo node can be marked as an offline friend.
  • the target user is defined as a seed user node in a three-dimensional space model, and other users on the social platform are defined as the target user.
  • a neighbor node of the seed user defining a pseudo node around the seed user; wherein the pseudo node is defined to be in mutual interest with the seed user node, and is defined as being on the social platform with the seed user Other online friends are not associated; then, the tightness between the seed user node and the neighbor nodes corresponding to other users on the social platform is calculated by a preset random walk algorithm, and the seed user node is closely related to the pseudo node Determining whether the closeness of the neighbor node and the seed user node calculated by the random walk algorithm is greater than the closeness of the seed user node and the pseudo node; and calculating the random walk algorithm The closeness of the seed node of the seed user node to the seed user node is greater than the species The neighbor node of the tightness of
  • a flowchart of a method for the real-world relationship matching method 102 of the social platform user in the second embodiment of the present application is proposed based on the real-world relationship matching method 100 of the social platform user in the first embodiment of the present application.
  • the steps S10-S40 in the real-world relationship matching method 102 of the social platform user are the same as those in the first embodiment, and are not described herein again; the difference lies in the reality of the social platform user.
  • the relationship matching method 102 can also include:
  • step S50 the neighbor nodes that have been determined to be the offline friend relationship are defined as a set in the three-dimensional space model, and the neighbor nodes corresponding to other users on the social platform are calculated by the preset random walk algorithm. The closeness between them, and the closeness of the set with the pseudo node;
  • Step S60 determining whether the closeness of the set neighbor node and the set calculated by the random walk algorithm is greater than the closeness of the set and the pseudo node;
  • Step S70 determining, by the random walk algorithm, the neighbor nodes of the set and the neighbor nodes whose closeness is greater than the closeness of the set and the pseudo node is determined to be offline with the target user. Friendship.
  • the neighbor node of the offline friend relationship when the neighbor node of the offline friend relationship has been determined, an iterative flag is performed, and the neighbor node that has been determined to be the offline friend relationship is defined as a set in the three-dimensional space model, by using the Presetting a random walk algorithm to calculate a closeness between the set and a neighbor node corresponding to other users on the social platform, and a closeness of the set to the pseudo node to determine other neighbors in the social network Whether the node is an offline friend relationship with the target user.
  • the neighbor node that has been determined to be the offline friend relationship is defined as a set in the three-dimensional space model, and the next neighbor node method of the offline friend relationship is iteratively calculated, thereby further increasing the real relationship matching method of the social platform user. Accuracy and recognition rate.
  • FIG. 3 a flowchart of a method for the real-world relationship matching method 103 of the social platform user in the third embodiment of the present application is proposed based on the real-world relationship matching method 102 of the social platform user in the second embodiment of the present application.
  • the steps S10-S70 in the real-world relationship matching method 103 of the social platform user are the same as those in the second embodiment, and are not described herein again; the difference is in the reality of the social platform user.
  • the relationship matching method 103 may further include:
  • Step S80 determining whether there is a new neighbor node determined to be an offline friend relationship with the target user
  • Step S91 when there is a new neighbor node determined to be a offline friend relationship with the target user, returning the neighbor node that has been determined to be an offline friend relationship is defined as a set in the three-dimensional space model, and
  • the preset random walk algorithm calculates a tightness between the set and neighbor nodes corresponding to other users on the social platform, and a step of tightness of the set and the pseudo node;
  • Step S92 when there is no new neighbor node determined to be an offline friend relationship with the target user, the process ends.
  • a neighbor node of the offline friend relationship is defined as a set in the three-dimensional space model, and the tightness between the set and neighbor nodes corresponding to other users on the social platform is calculated by the preset random walk algorithm, and The set is closely related to the pseudo node to determine whether other neighbor nodes in the social network and the target user are offline friend relationships.
  • the neighbor node that has been determined to be the offline friend relationship is defined as a set in the three-dimensional space model, and the next neighbor node method of the offline friend relationship is iteratively calculated, thereby further increasing the real relationship matching method of the social platform user. Accuracy and recognition rate.
  • the random walk algorithm may include at least one of the following weight parameters: information conveyable parameters, relationship exclusive parameters, and social circle viscous parameters.
  • the information conveyable parameter represents a probability that information can be transmitted from one node to another.
  • the relationship exclusive parameter represents a probability probability that a user maintains a certain social relationship on the line and offline in the domain user.
  • the proposed relationship exclusivity does not mean direct exclusivity of the relationship; for example, A and B maintain a certain social relationship and do not directly exclude A and C from maintaining a certain social relationship, only because of A's energy. Limited, A and B maintain a certain social relationship will indirectly reduce the possibility of A and others maintain a certain social relationship. Therefore, the relationship exclusivity parameter can be used to calculate the probability of a user maintaining a certain social relationship with the domain user online and offline.
  • the social circle sticky parameter is a representation probability of an offline friend relationship based on a common friend relationship.
  • the social circle sticky parameter can be used to discover those users who have no direct two-way attention with the target user but have strong association with the target user's offline friends, and such users and target users are likely to be offline friends.
  • the weight of the information conveyable parameter is less than the weight of the relationship exclusive parameter, and the weight of the relationship exclusive parameter is less than the weight of the social circle sticky parameter.
  • the random walk algorithm may further include other weight parameters, and the weight ratio between the weight parameters may be set as needed.
  • the three-dimensional spatial model is a three-dimensional topological model or a three-dimensional spherical model.
  • FIG. 4 a flowchart of a method for the social network user's real relationship matching method 104 in the fourth embodiment of the present application is proposed based on the real relationship matching method 100 of the social platform user in the first embodiment of the present application.
  • the steps S10-S40 in the real-world relationship matching method 104 of the social platform user are the same as those in the first embodiment, and are not described herein again; the difference is in the reality of the social platform user.
  • the relationship matching method 104 can also include:
  • Step S11 determining that there is a mutual relationship between the new neighbor node and the seed user node
  • step S12 when it is determined that there is a mutual interest relationship between the new neighboring node and the seed user node, the calculation is performed between a neighboring node corresponding to another user on the social platform by using a preset random walk algorithm. Tightness, and the step of tightness of the seed user node to the pseudo node.
  • the fourth embodiment it is determined that there is a mutual interest relationship between the user and the seed user node on the social platform, and when it is determined that there is a mutual relationship between the new neighbor node and the seed user node, Triggering the iterative calculation of the actual relationship matching of the new round of social platform users, so as to mine the real relationship of the social platform users in real time.
  • the real-life relationship matching method of the social platform user in the above embodiment can accurately mine the real relationship of the online user from the hundreds of millions of users in the social platform, and the realistic relationship matching method of the social platform user can be applied to the financial product field. , public security monitoring areas, etc.
  • the internal data of a financial company is generally a single user data, and there is no correlation between the user and the user.
  • the degree of relationship between people is very important. For example, in the risk control model, assuming that a user borrows, there is no bad record in itself, but his family or close friend has a bad credit record. Then, when it comes to credit evaluation, it should be more careful.
  • the financial company establishes the user's close friend matching network according to the social account information in the external data user data, such as Sina Weibo, WeChat, etc., and can record the bad credit records of the user, his family, acquaintances, and friends on the matching network. To better control risk and prevent potential losses.
  • FIG. 5 is a block diagram of a data processing apparatus 200 according to an embodiment of the present application.
  • the data processing apparatus 200 includes a memory 201, a processor 202, and a computer program stored on the memory and operable on the processor 202.
  • the processor 202 implements the program to implement the following steps:
  • Step S10 providing a target user of a social platform, defining the target user as a seed user node in a three-dimensional space model, and defining other users on the social platform as neighbor nodes of the seed user, around the seed user Defining a pseudo node; wherein the pseudo node is defined to be in contact with the seed user node and is defined as not associated with other online friends of the seed user on the social platform;
  • Step S20 calculating, by a preset random walk algorithm, a closeness between the seed user node and a neighbor node corresponding to other users on the social platform, and a closeness between the seed user node and the pseudo node;
  • Step S30 Determine whether the closeness of the neighbor node and the seed user node calculated by the random walk algorithm is greater than the closeness of the seed user node and the pseudo node.
  • Step S40 determining, by using the random walk algorithm, the neighbor node of the seed user node that is closer to the seed user node than the neighbor node of the seed user node and the pseudo node is determined to be
  • the target user is an offline friend relationship.
  • the data processing apparatus defines the target user as a seed user node in a three-dimensional model by defining a target user of a social platform, and defines other users on the social platform as the seed user. a neighbor node defining a pseudo node around the seed user; wherein the pseudo node is defined to be in mutual interest with the seed user node, and is defined as other online on the social platform with the seed user The friend has no association; then, the tightness between the seed user node and the neighbor node corresponding to other users on the social platform is calculated by a preset random walk algorithm, and the closeness of the seed user node and the pseudo node; Determining whether the closeness of the neighbor node and the seed user node calculated by the random walk algorithm is greater than the closeness of the seed user node and the pseudo node; and the calculating the random walk algorithm The closeness of the seed user node to the seed user node is greater than the seed user node
  • the neighbor node with the closeness of the pseudo node is determined to be an offline friend relationship
  • the data processing device 200 may be an electronic product having a data processing function such as a server, a computer, a portable computer device, a mobile phone, or a tablet computer.
  • processor 202 when the processor 202 executes the program, the following steps may be implemented:
  • step S50 the neighbor nodes that have been determined to be the offline friend relationship are defined as a set in the three-dimensional space model, and the neighbor nodes corresponding to other users on the social platform are calculated by the preset random walk algorithm. The closeness between them, and the closeness of the set with the pseudo node;
  • Step S60 determining whether the closeness of the set neighbor node and the set calculated by the random walk algorithm is greater than the closeness of the set and the pseudo node;
  • Step S70 determining, by the random walk algorithm, the neighbor nodes of the set and the neighbor nodes whose closeness is greater than the closeness of the set and the pseudo node is determined to be offline with the target user. Friendship.
  • processor 202 executes the program, the following steps may also be implemented:
  • Step S80 determining whether there is a new neighbor node determined to be an offline friend relationship with the target user
  • Step S91 when there is a new neighbor node determined to be a offline friend relationship with the target user, returning the neighbor node that has been determined to be an offline friend relationship is defined as a set in the three-dimensional space model, and
  • the preset random walk algorithm calculates a tightness between the set and neighbor nodes corresponding to other users on the social platform, and a step of tightness of the set and the pseudo node;
  • Step S92 when there is no new neighbor node determined to be an offline friend relationship with the target user, the process ends.
  • the random walk algorithm may include at least one of the following weight parameters: information conveyable parameters, relationship exclusive parameters, and social circle viscous parameters.
  • the weight of the information conveyable parameter is less than the weight of the relationship exclusive parameter, and the weight of the relationship exclusive parameter is less than the weight of the social circle sticky parameter.
  • the random walk algorithm may further include other weight parameters, and the weight ratio between the weight parameters may be set as needed.
  • the three-dimensional spatial model is a three-dimensional topological model or a three-dimensional spherical model.
  • processor 202 executes the program, the following steps may also be implemented:
  • Step S11 determining that there is a mutual relationship between the new neighbor node and the seed user node
  • step S12 when it is determined that there is a mutual interest relationship between the new neighboring node and the seed user node, the calculation is performed between a neighboring node corresponding to another user on the social platform by using a preset random walk algorithm. Tightness, and the step of tightness of the seed user node to the pseudo node.
  • FIG. 6 is a schematic structural diagram of a function module of the data processing apparatus 200 according to an embodiment of the present application.
  • the data processing device 200 includes:
  • a model node definition module 210 which provides a target user of a social platform, in the three-dimensional space model, the target user is defined as a seed user node, and other users on the social platform are defined as neighbor nodes of the seed user, Defining a pseudo node around the seed user; wherein the pseudo node is defined to be in mutual interest with the seed user node and is defined as not associated with other online friends of the seed user on the social platform;
  • the tightness calculation module 220 calculates the closeness between the seed user node and the neighbor node corresponding to other users on the social platform by using a preset random walk algorithm, and the closeness of the seed user node and the pseudo node ;
  • the tightness judging module 230 determines whether the closeness of the neighbor node and the seed user node calculated by the random walk algorithm is greater than the closeness of the seed user node and the pseudo node.
  • the offline relationship determining module 240 the neighboring node of the seed user node calculated by the random walk algorithm is closer to the seed user node than the seed user node and the close relationship of the pseudo node The node is determined to be an offline friend relationship with the target user.
  • the data processing device 200 may be an electronic product having a data processing function such as a server, a computer, a portable computer device, a mobile phone, or a tablet computer.
  • the data processing apparatus 200 further includes:
  • the tightness iterative calculation module 250 defines a neighbor node that has been determined to be an offline friend relationship as a set in the three-dimensional space model, and calculates the set and other users on the social platform by using the preset random walk algorithm. The closeness between the corresponding neighbor nodes and the closeness of the set with the pseudo node;
  • the tightness judging module 230 is further configured to determine whether the closeness of the set neighbor node and the set calculated by the random walk algorithm is greater than the closeness of the set and the pseudo node;
  • the offline relationship determining module 240 is further configured to determine, by the neighbor node of the set calculated by the random walk algorithm, a neighbor node whose closeness is greater than a tightness of the set and the pseudo node. It is an offline friend relationship with the target user.
  • the data processing apparatus 200 further includes:
  • the iterative calculation triggering module 260 determines whether there is a new neighbor node determined to be an offline friend relationship with the target user
  • the tightness iteration calculation module 250 is further configured to: in the three-dimensional space model, the neighboring node that has been determined to be an offline friend relationship, when the existing neighbor node is determined to be an offline friend relationship with the target user. Defined as a set, the tightness between the set and neighbor nodes corresponding to other users on the social platform, and the closeness of the set with the pseudo node are calculated by the preset random walk algorithm.
  • the random walk algorithm may include at least one of the following weight parameters: information conveyable parameters, relationship exclusive parameters, and social circle viscous parameters.
  • the weight of the information conveyable parameter is less than the weight of the relationship exclusive parameter, and the weight of the relationship exclusive parameter is less than the weight of the social circle sticky parameter.
  • the random walk algorithm may further include other weight parameters, and the weight ratio between the weight parameters may be set as needed.
  • the three-dimensional spatial model is a three-dimensional topological model or a three-dimensional spherical model.
  • the data processing apparatus 200 further includes:
  • the seed user attention detection module 270 determines that there is a mutual relationship between the new neighbor node and the seed user node;
  • the tightness calculation module 220 calculates the neighbor node corresponding to the other users on the social platform by using the preset random walking algorithm. The closeness between the two, and the closeness of the seed user node to the pseudo node.
  • the application further provides a computer readable storage medium having stored thereon a computer program, the program being executable by the processor to implement the steps of the social network user's real relationship matching method according to any of the above.

Abstract

本申请公开了一种社交平台用户的现实关系匹配方法,包括步骤:定义一个种子用户节点,在所述种子用户周围定义一个和所述种子用户节点相互关注、与该种子用户的其他在线朋友无关联伪节点;通过预设的随机行走算法计算所述种子用户节点与其他用户对应的邻居节点之间的紧密度,以及所述种子用户节点与所述伪节点的紧密度;将所述通过随机行走算法计算出的所述种子用户节点的邻居节点中与该种子用户节点的紧密度大于所述种子用户节点与所述伪节点的紧密度的邻居节点确定为与所述目标用户是线下朋友关系。本申请还提供一种数据处理装置及计算机可读存储介质。

Description

社交平台用户的现实关系匹配方法、装置及可读存储介质 技术领域
本申请涉及计算机网络技术领域,尤其涉及一种社交平台用户的现实关系匹配方法、数据处理装置及计算机可读存储介质。
背景技术
随着互联网的发展,各类社交平台的流行,如何从上亿的海量用户中准确的挖掘出在线用户的现实关系,例如朋友、熟人和家人等亲密关系成为了一个重要的课题。
传统的社交网络分析中,通常做法是根据网络结构进行节点间距离的衡量,或者用聚类方法找到团簇来划分节点,在社交拓扑网络结构中通过不同的算法来计算用户之间的相似度,从而判断用户之间的关系。但是,由于社交网络特性,不少节点间虽然距离近,但是他们可能只是线上好友,线下或真实生活中根本没有见过面。
而在例如金融产品领域、公共安全监控领域等应用中,正确地计算和识别在线用户在现实生活中的亲密关系才是最迫切需要的。
申请内容
本申请的主要目的在于提供一种在线社交用户的现实关系匹配方法、数据处理装置及计算机可读存储介质,旨在解决如何准确地计算和识别在线用户在现实生活中的线下朋友关系的技术问题。
为实现上述目的,本申请提供的一种社交平台用户的现实关系匹配方法,包括步骤:
提供一社交平台的目标用户,在三维空间模型中将所述目标用户定义为种子用户节点,将该社交平台上的其他用户定义为该种子用户的邻居节点,在所述种子用户周围定义一个伪节点;其中,该伪节点被定义为和所述种子用户节点相互关注,以及被定义为与该种子用户的在所述社交平台上的其他在线朋友无关联;
通过预设的随机行走算法计算所述种子用户节点与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述种子用户节点与所述伪节点的紧密度;
判断所述通过随机行走算法计算出的所述邻居节点与该种子用户节点的紧密度是否大于所述种子用户节点与所述伪节点的紧密度;
将所述通过随机行走算法计算出的所述种子用户节点的邻居节点中与该种子用户节点的紧密度大于所述种子用户节点与所述伪节点的紧密度的邻居节点确定为与所述目标用户是线下朋友关系。
本申请还提供一种数据处理装置,包括:
模型节点定义模块,提供一社交平台的目标用户,在三维空间模型中将所述目标用户定义为种子用户节点,将该社交平台上的其他用户定义为该种子用户的邻居节点,在所述种子用户周围定义一个伪节点;其中,定义该伪节点和种子用户相互关注,以及与种子用户的其他在线朋友无关联;
紧密度计算模块,通过预设的随机行走算法计算所述种子用户节点与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述种子用户节点与所述伪节点的紧密度;
紧密度判断模块,判断所述通过随机行走算法计算出的所述种子用户节点的邻居节点中与该种子用户节点的紧密度是否大于所述种子用户节点与所述伪节点的紧密度;
线下关系判断模块,将所述通过随机行走算法计算出的所述种子用户节点的邻居节点中与该种子用户节点的紧密度大于所述种子用户节点与所述伪节点的紧密度的邻居节点确定为与所述目标用户是线下朋友关系。
本申请还提供一种数据处理装置,包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序,所述处理器执行所述程序时实现上述任一项所述的社交平台用户的现实关系匹配方法的步骤。
本申请还提供一种计算机可读存储介质,其上存储有计算机程序,该程序被处理器执行时实现上述任一项所述的社交平台用户的现实关系匹配方法的步骤。
在本申请中,通过提供一社交平台的目标用户,在三维空间模型中将所述目标用户定义为种子用户节点,将该社交平台上的其他用户定义为该种子用户的邻居节点,在所述种子用户周围定义一个伪节点;其中,该伪节点被定义为和所述种子用户节点相互关注,以及被定义为与该种子用户的在所述社交平台上的其他 在线朋友无关联;然后通过预设的随机行走算法计算所述种子用户节点与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述种子用户节点与所述伪节点的紧密度;判断所述通过随机行走算法计算出的所述邻居节点与该种子用户节点的紧密度是否大于所述种子用户节点与所述伪节点的紧密度;将所述通过随机行走算法计算出的所述种子用户节点的邻居节点中与该种子用户节点的紧密度大于所述种子用户节点与所述伪节点的紧密度的邻居节点确定为与所述目标用户是线下朋友关系;有效地解决了仅仅进行用户的个人信息分析,通常只能分析出线上好友的缺点;同时,解决了通常的仅仅根据网络结构进行节点间距离,或者用聚类方法找到团簇来划分节点,在社交拓扑网络结构中通过不同的算法来计算用户之间的相似度,从而判断用户之间的关系的在线社交用户的现实关系匹配方式中,智能反应线上好友关系,而不能真实的反应线下朋友关系的缺点。
附图说明
图1为本申请第一实施方式中的社交平台用户的现实关系匹配方法的流程图;
图2为本申请第二实施方式中的社交平台用户的现实关系匹配方法的流程图;
图3为本申请第三实施方式中的社交平台用户的现实关系匹配方法的流程图;
图4为本申请第四实施方式中的社交平台用户的现实关系匹配方法的流程图;
图5为本申请一实施方式中的数据处理装置的硬件模块结构示意图。
图6为本申请一实施方式中的数据处理装置的功能模块结构示意图。
本申请目的的实现、功能特点及优点将结合实施例,参照附图做进一步说明。
具体实施方式
应当理解,此处所描述的实施例仅仅用以解释本申请,并不用于限定本申请。
请参照图1,图1为本申请第一实施方式中的社交平台用户的现实关系匹配方法100的方法流程图,其中,所述社交平台用户的现实关系匹配方法100包括如下步骤:
步骤S10,提供一社交平台的目标用户,在三维空间模型中将所述目标用户定义为种子用户节点,将该社交平台上的其他用户定义为该种子用户的邻居节点,在所述种子用户周围定义一个伪节点;其中,该伪节点被定义为和所述种子用户节点相互关注,以及被定义为与该种子用户的在所述社交平台上的其他在线朋友无关联。
其中,所述社交平台上的目标用户可以是同一个平台上的注册的用户,也可以是跨社交平台上的不同社交平台的用户。所述社交平台可以是,互联网中的各类社交软件或者社交网站等,在此不做具体限制。
步骤S20,通过预设的随机行走算法计算所述种子用户节点与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述种子用户节点与所述伪节点的紧密度。
所述随机行走算法(random walk)是一个算法,通常用于计算在一个一维/二维/三维网络中,从一个节点随机走到另一个节点的概率,因此可以用于度量各节点之间的紧密度。
所述伪节点是一种假想的与所述种子用户相互关注的一个假想的节点,相当于一个预设的值。在本实施例中,所述伪节点是出来区分线下用户的,该伪节点被定义为和所述种子用户节点相互关注,以及被定义为与该种子用户的在所述社交平台上的其他在线朋友无关联;这里伪节点模拟了一个与种子用户因巧合相互关注的线上朋友,因此,用种子用户与这个伪节点的紧密度可以作为阀值区分线下朋友。
在本实施例中,所述随机行走算法中,所述三维空间模型中的从一个节点行走到另一个节点的概率可以是随机的也可以是不随机具有一定权重的概率。
在一实施例中,所述随机行走算法的目标函数可以设定为:其中,假设所述三维空间模型G(V,E)是一个连通图,有n个节点和m条边,在t=0时刻从种子用户的节点v 0出发,在t时刻到达v t点,然后以概率1/k(v t)移动到v t的任一邻居节点上,k(v t)是节点v t的邻居节点的数目;
那么从节点v 0移动到v t节点的概率P t为:
P t(i)=Pr ob·{v t=i};
其中,prob函数为数值落在指定区间内的对应概率的函数,ⅰ∈(1,n)。
步骤S30,判断所述通过随机行走算法计算出的所述邻居节点与该种子用户 节点的紧密度是否大于所述种子用户节点与所述伪节点的紧密度;
步骤S40,将所述通过随机行走算法计算出的所述种子用户节点的邻居节点中与该种子用户节点的紧密度大于所述种子用户节点与所述伪节点的紧密度的邻居节点确定为与所述目标用户是线下朋友关系。
在本实施例中,所述伪节点是一个用来区分线下朋友和非线下朋友的假想节点,该伪节点仅与目标用户相互关注,因此基于随机行走算法计算出的所有邻居节点与该种子用户节点的紧密度大于所述伪节点的用户可以被标注为线下朋友。
在所述社交平台用户的现实关系匹配方法100中,通过提供一社交平台的目标用户,在三维空间模型中将所述目标用户定义为种子用户节点,将该社交平台上的其他用户定义为该种子用户的邻居节点,在所述种子用户周围定义一个伪节点;其中,该伪节点被定义为和所述种子用户节点相互关注,以及被定义为与该种子用户的在所述社交平台上的其他在线朋友无关联;然后通过预设的随机行走算法计算所述种子用户节点与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述种子用户节点与所述伪节点的紧密度;判断所述通过随机行走算法计算出的所述邻居节点与该种子用户节点的紧密度是否大于所述种子用户节点与所述伪节点的紧密度;将所述通过随机行走算法计算出的所述种子用户节点的邻居节点中与该种子用户节点的紧密度大于所述种子用户节点与所述伪节点的紧密度的邻居节点确定为与所述目标用户是线下朋友关系;有效地解决了仅仅进行用户的个人信息分析,通常只能分析出线上好友的缺点;同时,解决了通常的仅仅根据网络结构进行节点间距离,或者用聚类方法找到团簇来划分节点,在社交拓扑网络结构中通过不同的算法来计算用户之间的相似度,从而判断用户之间的关系的在线社交用户的现实关系匹配方式中,智能反应线上好友关系,而不能真实的反应线下朋友关系的缺点。
请一并参考图2,为基于本申请第一实施例中的社交平台用户的现实关系匹配方法100提出本申请第二实施例中的社交平台用户的现实关系匹配方法102的方法流程图。
在本实施例中,所述社交平台用户的现实关系匹配方法102中的所述步骤S10~S40均与第一实施例相同,在此不再赘述;其不同在于,所述社交平台用户的现实关系匹配方法102还可以包括:
步骤S50,将已经确定为线下朋友关系的邻居节点在所述三维空间模型中定 义为一个集合,通过所述预设的随机行走算法计算所述集合与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述集合与所述伪节点的紧密度;
步骤S60,判断所述通过随机行走算法计算出的所述集合的邻居节点与该集合的紧密度是否大于所述集合与所述伪节点的紧密度;
步骤S70,将所述通过随机行走算法计算出的所述集合的邻居节点与该集合的紧密度大于所述集合与所述伪节点的紧密度的邻居节点确定为与所述目标用户是线下朋友关系。
在该第二实施例中,在已经确定有线下朋友关系的邻居节点时,进行迭代标记,将已经确定为线下朋友关系的邻居节点在所述三维空间模型中定义为一个集合,通过所述预设的随机行走算法计算所述集合与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述集合与所述伪节点的紧密度,来判断所述社交网络中的其他邻居节点与所述目标用户是否是线下朋友关系。采用将已经确定为线下朋友关系的邻居节点在所述三维空间模型中定义为一个集合,迭代计算下一个为线下朋友关系的邻居节点方式,进一步增加了社交平台用户的现实关系匹配方法的准确率和识别率。
请一并参考图3,为基于本申请第二实施例中的社交平台用户的现实关系匹配方法102提出本申请第三实施例中的社交平台用户的现实关系匹配方法103的方法流程图。
在本实施例中,所述社交平台用户的现实关系匹配方法103中的所述步骤S10~S70均与第二实施例相同,在此不再赘述;其不同在于,所述社交平台用户的现实关系匹配方法103还可以包括:
步骤S80,判断是否存在新的邻居节点确定为与所述目标用户是线下朋友关系;
步骤S91,在存在新的邻居节点确定为与所述目标用户是线下朋友关系时,返回所述将已经确定为线下朋友关系的邻居节点在所述三维空间模型中定义为一个集合,通过所述预设的随机行走算法计算所述集合与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述集合与所述伪节点的紧密度的步骤;
步骤S92,在不存在新的邻居节点确定为与所述目标用户是线下朋友关系时,则结束。
在该第三实施例中,通过主动判断是否存在新的邻居节点确定为与所述目标 用户是线下朋友关系,在已经确定有线下朋友关系的邻居节点时,进行迭代标记,将已经确定为线下朋友关系的邻居节点在所述三维空间模型中定义为一个集合,通过所述预设的随机行走算法计算所述集合与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述集合与所述伪节点的紧密度,来判断所述社交网络中的其他邻居节点与所述目标用户是否是线下朋友关系。采用将已经确定为线下朋友关系的邻居节点在所述三维空间模型中定义为一个集合,迭代计算下一个为线下朋友关系的邻居节点方式,进一步增加了社交平台用户的现实关系匹配方法的准确率和识别率。
在一实施例中,所述随机行走算法可以包括以下权重参数中的至少一种:信息可传达参数,关系排他性参数以及社交圈粘性参数。
其中,所述信息可传达参数代表的是信息从一个节点可以传输到另一个节点的概率。
在真实社会中,朋友的一个特点是经常沟通,而这意味着信息能够在朋友之间传递,我们称这为信息相互可达性。在微博等社交平台中关注关系是单向的,这意味着A的微博能够传递到B,而B的微博不一定能够传递到A。根据相关统计结果,微博中只有大约22\%的关系是双向(相互关注)的。因此,对于识别线下朋友,单向关注关系是很弱的证据,而如果仅考虑双向关注,又会排除太多的用户;因此,信息相互可达性是识别线下朋友的一个明显特征。
其中,所述关系排他性参数代表的是一个用户在在线上和线下于领域用户维持一定社会关系的可能性概率。
这是基于社会行为学中的观点:由于人类自身大脑处理能力的限制,人们维持稳定社会关系的数量是有上限的;而这一上限通常被称为Dunbar数,大概介于100和230之间。当然,在本实施例中,所提出的关系排他性并非意味着关系的直接排他性;例如,A与B维持了一定的社会关系并不直接排斥A与C维持一定的社会关系,只是因为A的精力有限,A与B维持了一定的社会关系会间接降低A与其他人维持一定社会关系的可能性。因此,关系排他性参数可以用来计算一个用户在在线上和线下于领域用户维持一定社会关系的可能性概率。
其中,所述社交圈粘性参数是一种基于共同朋友关系的线下朋友关系的表征概率。
在人类社会中的个人并非单独地处于社会之中,而是非常自然的以某些方式 参与一些社会交际圈,并与这些社会交际圈中的人维持稳定社会关系,而这些社会交际圈中的其他人之间也以同样的方式维持着稳定的社会关系。举例来说,这样的社会交际圈有学校同学关系圈,公司同事关系圈,家族亲戚关系圈等等。两个在相同社会交际圈中的人自然地拥有一些共同的朋友,这些共同朋友越多,这两个人是朋友的概率就越大,这种情形被称之为社交圈粘性。
因此,所述社交圈粘性参数可以用于发现那些与目标用户没有直接双向关注却与目标用户的线下朋友有很强关联的用户,而这样的用户与目标用户很有可能也是线下朋友。
可选地,在一实施例中,所述信息可传达参数的权重小于所述关系排他性参数的权重,所述关系排他性参数的权重小于社交圈粘性参数的权重。
可以理解的是,所述随机行走算法还可以包括其他的权重参数,各个权重参数之间的权重比例可以根据需要进行设置。
进一步地,在一实施例中,所述三维空间模型为三维拓扑模型或者三维球形模型。
请一并参考图4,为基于本申请第一实施例中的社交平台用户的现实关系匹配方法100提出本申请第四实施例中的社交平台用户的现实关系匹配方法104的方法流程图。
在本实施例中,所述社交平台用户的现实关系匹配方法104中的所述步骤S10~S40均与第一实施例相同,在此不再赘述;其不同在于,所述社交平台用户的现实关系匹配方法104还可以包括:
步骤S11,判断存在新的邻居节点与所述种子用户节点存在相互关注关系;
步骤S12,在确定存在新的邻居节点与所述种子用户节点存在相互关注关系时,进入通过预设的随机行走算法计算所述种子用户节点与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述种子用户节点与所述伪节点的紧密度的步骤。
在该第四实施例中,主动判断在所述社交平台上是否有的用户与所述种子用户节点存在相互关注关系,在确定存在新的邻居节点与所述种子用户节点存在相互关注关系时,触发新的一轮社交平台用户的现实关系匹配的迭代计算,从而实时的挖掘出社交平台用户的现实关系。
上述实施例中的社交平台用户的现实关系匹配方法可以从社交平台里上亿 的海量用户中准确的挖掘出在线用户的现实关系,所述社交平台用户的现实关系匹配方法可以应用于金融产品领域、公共安全监控领域等。
例如,金融公司内部数据一般是单个的用户数据,用户与用户之间是没有关联的。然而,在很多金融类产品应用中,人与人之间的关系程度是非常重要的。比方说,在风控模型中,假设某个用户来借款,其本身没有任何不良记录,然而他的家人或者密友曾经有不良信用记录。那么,在其信用评估时,就应该更仔细一些。金融公司根据外部数据用户数据中的社交账号信息,例如新浪微博、微信等,建立用户的亲密好友匹配网络,可以将用户及其家人、熟人、朋友的不良信用记录在该匹配网络上进行扩展,从而更好的进行风险控制,防止潜在损失。
请一并结合图5,为本申请一实施方式中的数据处理装置200的模块结构示意图。
所述数据处理装置200包括存储器201、处理器202及存储在存储器上并可在处理器202上运行的计算机程序,所述处理器202执行所述程序时实现如下的步骤:
步骤S10,提供一社交平台的目标用户,在三维空间模型中将所述目标用户定义为种子用户节点,将该社交平台上的其他用户定义为该种子用户的邻居节点,在所述种子用户周围定义一个伪节点;其中,该伪节点被定义为和所述种子用户节点相互关注,以及被定义为与该种子用户的在所述社交平台上的其他在线朋友无关联;
步骤S20,通过预设的随机行走算法计算所述种子用户节点与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述种子用户节点与所述伪节点的紧密度;
步骤S30,判断所述通过随机行走算法计算出的所述邻居节点与该种子用户节点的紧密度是否大于所述种子用户节点与所述伪节点的紧密度。
步骤S40,将所述通过随机行走算法计算出的所述种子用户节点的邻居节点中与该种子用户节点的紧密度大于所述种子用户节点与所述伪节点的紧密度的邻居节点确定为与所述目标用户是线下朋友关系。
在本实施方式中,所述数据处理装置,通过提供一社交平台的目标用户,在三维空间模型中将所述目标用户定义为种子用户节点,将该社交平台上的其他用户定义为该种子用户的邻居节点,在所述种子用户周围定义一个伪节点;其中, 该伪节点被定义为和所述种子用户节点相互关注,以及被定义为与该种子用户的在所述社交平台上的其他在线朋友无关联;然后通过预设的随机行走算法计算所述种子用户节点与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述种子用户节点与所述伪节点的紧密度;判断所述通过随机行走算法计算出的所述邻居节点与该种子用户节点的紧密度是否大于所述种子用户节点与所述伪节点的紧密度;将所述通过随机行走算法计算出的所述种子用户节点的邻居节点中与该种子用户节点的紧密度大于所述种子用户节点与所述伪节点的紧密度的邻居节点确定为与所述目标用户是线下朋友关系;有效地解决了仅仅进行用户的个人信息分析,通常只能分析出线上好友的缺点;同时,解决了通常的仅仅根据网络结构进行节点间距离,或者用聚类方法找到团簇来划分节点,在社交拓扑网络结构中通过不同的算法来计算用户之间的相似度,从而判断用户之间的关系的在线社交用户的现实关系匹配方式中,智能反应线上好友关系,而不能真实的反应线下朋友关系的缺点。
其中,该数据处理装置200可以是服务器,计算机、便携式计算机设备、手机、平板电脑等具备数据处理功能的电子产品。
在一实施方式中,所述处理器202执行所述程序时还可以实现如下的步骤:
步骤S50,将已经确定为线下朋友关系的邻居节点在所述三维空间模型中定义为一个集合,通过所述预设的随机行走算法计算所述集合与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述集合与所述伪节点的紧密度;
步骤S60,判断所述通过随机行走算法计算出的所述集合的邻居节点与该集合的紧密度是否大于所述集合与所述伪节点的紧密度;
步骤S70,将所述通过随机行走算法计算出的所述集合的邻居节点与该集合的紧密度大于所述集合与所述伪节点的紧密度的邻居节点确定为与所述目标用户是线下朋友关系。
进一步地,所述处理器202执行所述程序时还可以实现如下的步骤:
步骤S80,判断是否存在新的邻居节点确定为与所述目标用户是线下朋友关系;
步骤S91,在存在新的邻居节点确定为与所述目标用户是线下朋友关系时,返回所述将已经确定为线下朋友关系的邻居节点在所述三维空间模型中定义为一个集合,通过所述预设的随机行走算法计算所述集合与该社交平台上其他用户 对应的邻居节点之间的紧密度,以及所述集合与所述伪节点的紧密度的步骤;
步骤S92,在不存在新的邻居节点确定为与所述目标用户是线下朋友关系时,则结束。
在一实施例中,所述随机行走算法可以包括一下权重参数中的至少一种:信息可传达参数,关系排他性参数以及社交圈粘性参数。
可选地,在一实施例中,所述信息可传达参数的权重小于所述关系排他性参数的权重,所述关系排他性参数的权重小于社交圈粘性参数的权重。
可以理解的是,所述随机行走算法还可以包括其他的权重参数,各个权重参数之间的权重比例可以根据需要进行设置。
进一步地,在一实施例中,所述三维空间模型为三维拓扑模型或者三维球形模型。
进一步地,所述处理器202执行所述程序时还可以实现如下的步骤:
步骤S11,判断存在新的邻居节点与所述种子用户节点存在相互关注关系;
步骤S12,在确定存在新的邻居节点与所述种子用户节点存在相互关注关系时,进入通过预设的随机行走算法计算所述种子用户节点与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述种子用户节点与所述伪节点的紧密度的步骤。
请一并结合图6,为本申请一实施方式中的数据处理装置200的功能模块结构示意图。
所述数据处理装置200包括:
模型节点定义模块210,提供一社交平台的目标用户,在三维空间模型中将所述目标用户定义为种子用户节点,将该社交平台上的其他用户定义为该种子用户的邻居节点,在所述种子用户周围定义一个伪节点;其中,该伪节点被定义为和所述种子用户节点相互关注,以及被定义为与该种子用户的在所述社交平台上的其他在线朋友无关联;
紧密度计算模块220,通过预设的随机行走算法计算所述种子用户节点与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述种子用户节点与所述伪节点的紧密度;
紧密度判断模块230,判断所述通过随机行走算法计算出的所述邻居节点与该种子用户节点的紧密度是否大于所述种子用户节点与所述伪节点的紧密度。
线下关系判断模块240,将所述通过随机行走算法计算出的所述种子用户节点的邻居节点中与该种子用户节点的紧密度大于所述种子用户节点与所述伪节点的紧密度的邻居节点确定为与所述目标用户是线下朋友关系。
其中,该数据处理装置200可以是服务器,计算机、便携式计算机设备、手机、平板电脑等具备数据处理功能的电子产品。
在一实施方式中,该数据处理装置200还包括:
紧密度迭代计算模块250,将已经确定为线下朋友关系的邻居节点在所述三维空间模型中定义为一个集合,通过所述预设的随机行走算法计算所述集合与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述集合与所述伪节点的紧密度;
所述紧密度判断模块230,还用于判断所述通过随机行走算法计算出的所述集合的邻居节点与该集合的紧密度是否大于所述集合与所述伪节点的紧密度;
所述线下关系判断模块240,还用于将所述通过随机行走算法计算出的所述集合的邻居节点与该集合的紧密度大于所述集合与所述伪节点的紧密度的邻居节点确定为与所述目标用户是线下朋友关系。
进一步地,该数据处理装置200还包括:
迭代计算触发模块260,判断是否存在新的邻居节点确定为与所述目标用户是线下朋友关系;
在存在新的邻居节点确定为与所述目标用户是线下朋友关系时,所述紧密度迭代计算模块250,还用于将已经确定为线下朋友关系的邻居节点在所述三维空间模型中定义为一个集合,通过所述预设的随机行走算法计算所述集合与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述集合与所述伪节点的紧密度。
在一实施例中,所述随机行走算法可以包括一下权重参数中的至少一种:信息可传达参数,关系排他性参数以及社交圈粘性参数。
可选地,在一实施例中,所述信息可传达参数的权重小于所述关系排他性参数的权重,所述关系排他性参数的权重小于社交圈粘性参数的权重。
可以理解的是,所述随机行走算法还可以包括其他的权重参数,各个权重参数之间的权重比例可以根据需要进行设置。
进一步地,在一实施例中,所述三维空间模型为三维拓扑模型或者三维球形 模型。
进一步地,该数据处理装置200还包括:
种子用户关注侦测模块270,判断存在新的邻居节点与所述种子用户节点存在相互关注关系;
在确定存在新的邻居节点与所述种子用户节点存在相互关注关系时,所述紧密度计算模块220通过预设的随机行走算法计算所述种子用户节点与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述种子用户节点与所述伪节点的紧密度。
本申请还提供一种计算机可读存储介质,其上存储有计算机程序,该程序被处理器执行时可以实现如上述任一项所述的社交平台用户的现实关系匹配方法的步骤。
可以理解的是,在本说明书的描述中,参考术语“一实施例”、“另一实施例”、“其他实施例”、或“第一实施例~第N实施例”等的描述意指结合该实施例或示例描述的特征、结构、材料或者特点包含于本申请的至少一个实施例或示例中。在本说明书中,对上述术语的示意性表述不一定指的是相同的实施例或示例。而且,描述的特征、结构、材料或者特点可以在任何的一个或多个实施例或示例中以合适的方式结合。
可以理解的是,以上仅为本申请的可选实施例,并非因此限制本申请的专利范围,凡是利用本申请说明书及附图内容所作的等效结构或等效流程变换,或直接或间接运用在其他相关的技术领域,均同理包括在本申请的专利保护范围内。

Claims (28)

  1. 一种社交平台用户的现实关系匹配方法,其中,包括步骤:
    确定社交平台的目标用户,在三维空间模型中将所述目标用户定义为种子用户节点,将该社交平台上的其他用户定义为该种子用户的邻居节点,在所述种子用户周围定义一个伪节点;其中,该伪节点被定义为和所述种子用户节点相互关注,以及被定义为与该种子用户的在所述社交平台上的其他在线朋友无关联;
    通过预设的随机行走算法计算所述种子用户节点与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述种子用户节点与所述伪节点的紧密度;
    判断所述通过随机行走算法计算出的所述邻居节点与该种子用户节点的紧密度是否大于所述种子用户节点与所述伪节点的紧密度;
    将所述通过随机行走算法计算出的所述种子用户节点的邻居节点中与该种子用户节点的紧密度大于所述种子用户节点与所述伪节点的紧密度的邻居节点确定为与所述目标用户是线下朋友关系。
  2. 如权利要求1所述的社交平台用户的现实关系匹配方法,其中,还包括步骤:
    判断存在新的邻居节点与所述种子用户节点存在相互关注关系;
    在确定存在新的邻居节点与所述种子用户节点存在相互关注关系时,进入通过预设的随机行走算法计算所述种子用户节点与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述种子用户节点与所述伪节点的紧密度的步骤。
  3. 如权利要求1所述的社交平台用户的现实关系匹配方法,其中,还包括步骤:
    将已经确定为线下朋友关系的邻居节点在所述三维空间模型中定义为一个集合,通过所述预设的随机行走算法计算所述集合与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述集合与所述伪节点的紧密度;
    判断所述通过随机行走算法计算出的所述集合的邻居节点与该集合的紧密度是否大于所述集合与所述伪节点的紧密度;
    将所述通过随机行走算法计算出的所述集合的邻居节点与该集合的紧密度大于所述集合与所述伪节点的紧密度的邻居节点确定为与所述目标用户是线下朋友关系。
  4. 如权利要求3所述的社交平台用户的现实关系匹配方法,其中,还包括步骤:
    判断存在新的邻居节点与所述种子用户节点存在相互关注关系;
    在确定存在新的邻居节点与所述种子用户节点存在相互关注关系时,进入通过预设的随机行走算法计算所述种子用户节点与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述种子用户节点与所述伪节点的紧密度的步骤。
  5. 如权利要求4所述的社交平台用户的现实关系匹配方法,其中,所述随机行走算法包括以下权重参数中的至少一种:信息可传达参数,关系排他性参数以及社交圈粘性参数;所述信息可传达参数的权重小于所述关系排他性参数的权重,所述关系排他性参数的权重小于社交圈粘性参数的权重。
  6. 如权利要求5所述的社交平台用户的现实关系匹配方法,其中,所述三维空间模型为三维拓扑模型或者三维球形模型。
  7. 如权利要求3所述的社交平台用户的现实关系匹配方法,其中,所述随机行走算法包括以下权重参数中的至少一种:信息可传达参数,关系排他性参数以及社交圈粘性参数;所述信息可传达参数的权重小于所述关系排他性参数的权重,所述关系排他性参数的权重小于社交圈粘性参数的权重。
  8. 如权利要求3所述的社交平台用户的现实关系匹配方法,其中,所述三维空间模型为三维拓扑模型或者三维球形模型。
  9. 如权利要求3所述的社交平台用户的现实关系匹配方法,其中,还包括步骤:
    判断是否存在新的邻居节点确定为与所述目标用户是线下朋友关系;
    在存在新的邻居节点确定为与所述目标用户是线下朋友关系时,返回所述将已经确定为线下朋友关系的邻居节点在所述三维空间模型中定义为一个集合,通过所述预设的随机行走算法计算所述集合与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述集合与所述伪节点的紧密度的步骤;
    在不存在新的邻居节点确定为与所述目标用户是线下朋友关系时,则结束。
  10. 如权利要求9所述的社交平台用户的现实关系匹配方法,其中,还包括步骤:
    判断存在新的邻居节点与所述种子用户节点存在相互关注关系;
    在确定存在新的邻居节点与所述种子用户节点存在相互关注关系时,进入通 过预设的随机行走算法计算所述种子用户节点与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述种子用户节点与所述伪节点的紧密度的步骤。
  11. 如权利要求10所述的社交平台用户的现实关系匹配方法,其中,所述随机行走算法包括以下权重参数中的至少一种:信息可传达参数,关系排他性参数以及社交圈粘性参数;
  12. 如权利要求11所述的社交平台用户的现实关系匹配方法,其中,所述信息可传达参数的权重小于所述关系排他性参数的权重,所述关系排他性参数的权重小于社交圈粘性参数的权重。
  13. 如权利要求12所述的社交平台用户的现实关系匹配方法,其中,所述三维空间模型为三维拓扑模型或者三维球形模型。
  14. 如权利要求1所述的社交平台用户的现实关系匹配方法,其中,所述随机行走算法包括以下权重参数中的至少一种:信息可传达参数,关系排他性参数以及社交圈粘性参数。
  15. 如权利要求14所述的社交平台用户的现实关系匹配方法,其中,所述信息可传达参数的权重小于所述关系排他性参数的权重,所述关系排他性参数的权重小于社交圈粘性参数的权重。
  16. 如权利要求15所述的社交平台用户的现实关系匹配方法,其中,还包括步骤:
    判断存在新的邻居节点与所述种子用户节点存在相互关注关系;
    在确定存在新的邻居节点与所述种子用户节点存在相互关注关系时,进入通过预设的随机行走算法计算所述种子用户节点与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述种子用户节点与所述伪节点的紧密度的步骤。
  17. 如权利要求1所述的社交平台用户的现实关系匹配方法,其中,所述三维空间模型为三维拓扑模型或者三维球形模型。
  18. 一种数据处理装置,其中,包括:
    模型节点定义模块,提供一社交平台的目标用户,在三维空间模型中将所述目标用户定义为种子用户节点,将该社交平台上的其他用户定义为该种子用户的邻居节点,在所述种子用户周围定义一个伪节点;其中,定义该伪节点和种子用户相互关注,以及与种子用户的其他在线朋友无关联;
    紧密度计算模块,通过预设的随机行走算法计算所述种子用户节点与该社交 平台上其他用户对应的邻居节点之间的紧密度,以及所述种子用户节点与所述伪节点的紧密度;
    紧密度判断模块,判断所述通过随机行走算法计算出的所述种子用户节点的邻居节点中与该种子用户节点的紧密度是否大于所述种子用户节点与所述伪节点的紧密度;
    线下关系判断模块,将所述通过随机行走算法计算出的所述种子用户节点的邻居节点中与该种子用户节点的紧密度大于所述种子用户节点与所述伪节点的紧密度的邻居节点确定为与所述目标用户是线下朋友关系。
  19. 一种数据处理装置,包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序,其中,所述处理器执行所述程序时实现以下步骤:
    确定社交平台的目标用户,在三维空间模型中将所述目标用户定义为种子用户节点,将该社交平台上的其他用户定义为该种子用户的邻居节点,在所述种子用户周围定义一个伪节点;其中,该伪节点被定义为和所述种子用户节点相互关注,以及被定义为与该种子用户的在所述社交平台上的其他在线朋友无关联;
    通过预设的随机行走算法计算所述种子用户节点与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述种子用户节点与所述伪节点的紧密度;
    判断所述通过随机行走算法计算出的所述邻居节点与该种子用户节点的紧密度是否大于所述种子用户节点与所述伪节点的紧密度;
    将所述通过随机行走算法计算出的所述种子用户节点的邻居节点中与该种子用户节点的紧密度大于所述种子用户节点与所述伪节点的紧密度的邻居节点确定为与所述目标用户是线下朋友关系。
  20. 如权利要求19所述的数据处理装置,其中,所述处理器执行所述程序时还实现以下步骤:
    将已经确定为线下朋友关系的邻居节点在所述三维空间模型中定义为一个集合,通过所述预设的随机行走算法计算所述集合与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述集合与所述伪节点的紧密度;
    判断所述通过随机行走算法计算出的所述集合的邻居节点与该集合的紧密度是否大于所述集合与所述伪节点的紧密度;
    将所述通过随机行走算法计算出的所述集合的邻居节点与该集合的紧密度大于所述集合与所述伪节点的紧密度的邻居节点确定为与所述目标用户是线下 朋友关系。
  21. 如权利要求20所述的数据处理装置,其中,所述处理器执行所述程序时还实现以下步骤:
    判断是否存在新的邻居节点确定为与所述目标用户是线下朋友关系;
    在存在新的邻居节点确定为与所述目标用户是线下朋友关系时,返回所述将已经确定为线下朋友关系的邻居节点在所述三维空间模型中定义为一个集合,通过所述预设的随机行走算法计算所述集合与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述集合与所述伪节点的紧密度的步骤;
    在不存在新的邻居节点确定为与所述目标用户是线下朋友关系时,则结束。
  22. 如权利要求19所述的数据处理装置,其中,所述处理器执行所述程序时还实现以下步骤:
    判断存在新的邻居节点与所述种子用户节点存在相互关注关系;
    在确定存在新的邻居节点与所述种子用户节点存在相互关注关系时,进入通过预设的随机行走算法计算所述种子用户节点与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述种子用户节点与所述伪节点的紧密度的步骤。
  23. 如权利要求19所述的数据处理装置,其中,所述随机行走算法包括以下权重参数中的至少一种:信息可传达参数,关系排他性参数以及社交圈粘性参数;所述信息可传达参数的权重小于所述关系排他性参数的权重,所述关系排他性参数的权重小于社交圈粘性参数的权重;
    所述三维空间模型为三维拓扑模型或者三维球形模型。
  24. 一种计算机可读存储介质,其上存储有计算机程序,其中,该程序被处理器执行时实现步骤:
    确定社交平台的目标用户,在三维空间模型中将所述目标用户定义为种子用户节点,将该社交平台上的其他用户定义为该种子用户的邻居节点,在所述种子用户周围定义一个伪节点;其中,该伪节点被定义为和所述种子用户节点相互关注,以及被定义为与该种子用户的在所述社交平台上的其他在线朋友无关联;
    通过预设的随机行走算法计算所述种子用户节点与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述种子用户节点与所述伪节点的紧密度;
    判断所述通过随机行走算法计算出的所述邻居节点与该种子用户节点的紧密度是否大于所述种子用户节点与所述伪节点的紧密度;
    将所述通过随机行走算法计算出的所述种子用户节点的邻居节点中与该种子用户节点的紧密度大于所述种子用户节点与所述伪节点的紧密度的邻居节点确定为与所述目标用户是线下朋友关系。
  25. 如权利要求24所述的计算机可读存储介质,其中,该程序被处理器执行时还实现步骤:
    将已经确定为线下朋友关系的邻居节点在所述三维空间模型中定义为一个集合,通过所述预设的随机行走算法计算所述集合与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述集合与所述伪节点的紧密度;
    判断所述通过随机行走算法计算出的所述集合的邻居节点与该集合的紧密度是否大于所述集合与所述伪节点的紧密度;
    将所述通过随机行走算法计算出的所述集合的邻居节点与该集合的紧密度大于所述集合与所述伪节点的紧密度的邻居节点确定为与所述目标用户是线下朋友关系。
  26. 如权利要求25所述的计算机可读存储介质,其中,该程序被处理器执行时还实现步骤:
    判断是否存在新的邻居节点确定为与所述目标用户是线下朋友关系;
    在存在新的邻居节点确定为与所述目标用户是线下朋友关系时,返回所述将已经确定为线下朋友关系的邻居节点在所述三维空间模型中定义为一个集合,通过所述预设的随机行走算法计算所述集合与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述集合与所述伪节点的紧密度的步骤;
    在不存在新的邻居节点确定为与所述目标用户是线下朋友关系时,则结束。
  27. 如权利要求24所述的计算机可读存储介质,其中,该程序被处理器执行时还实现步骤:
    判断存在新的邻居节点与所述种子用户节点存在相互关注关系;
    在确定存在新的邻居节点与所述种子用户节点存在相互关注关系时,进入通过预设的随机行走算法计算所述种子用户节点与该社交平台上其他用户对应的邻居节点之间的紧密度,以及所述种子用户节点与所述伪节点的紧密度的步骤。
  28. 如权利要求24所述的计算机可读存储介质,其中,所述随机行走算法包括以下权重参数中的至少一种:信息可传达参数,关系排他性参数以及社交圈粘性参数;所述信息可传达参数的权重小于所述关系排他性参数的权重,所述关系 排他性参数的权重小于社交圈粘性参数的权重;
    所述三维空间模型为三维拓扑模型或者三维球形模型。
PCT/CN2018/075204 2017-10-25 2018-02-05 社交平台用户的现实关系匹配方法、装置及可读存储介质 WO2019080403A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201711013931.5 2017-10-25
CN201711013931.5A CN107908673B (zh) 2017-10-25 2017-10-25 社交平台用户的现实关系匹配方法、装置及可读存储介质

Publications (1)

Publication Number Publication Date
WO2019080403A1 true WO2019080403A1 (zh) 2019-05-02

Family

ID=61840893

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/075204 WO2019080403A1 (zh) 2017-10-25 2018-02-05 社交平台用户的现实关系匹配方法、装置及可读存储介质

Country Status (2)

Country Link
CN (1) CN107908673B (zh)
WO (1) WO2019080403A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10915919B2 (en) * 2017-01-03 2021-02-09 International Business Machines Corporation Topology-based online reward incentives

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110196955B (zh) * 2018-05-28 2023-07-07 腾讯科技(深圳)有限公司 信息处理方法、装置及存储介质
CN114596097B (zh) * 2022-05-10 2022-08-26 富算科技(上海)有限公司 用户识别方法、装置、电子设备和计算机可读存储介质

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103164416A (zh) * 2011-12-12 2013-06-19 阿里巴巴集团控股有限公司 一种用户关系的识别方法及设备
CN103544236A (zh) * 2013-10-07 2014-01-29 宁波芝立软件有限公司 一种通过确定未知关系人来推导亲缘关系方法
CN106959953A (zh) * 2016-01-08 2017-07-18 中国科学院声学研究所 一种基于统计特征的社交网络用户关系的计算方法

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103426042B (zh) * 2012-05-15 2016-04-13 腾讯科技(深圳)有限公司 社交网络的分组方法及系统

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103164416A (zh) * 2011-12-12 2013-06-19 阿里巴巴集团控股有限公司 一种用户关系的识别方法及设备
CN103544236A (zh) * 2013-10-07 2014-01-29 宁波芝立软件有限公司 一种通过确定未知关系人来推导亲缘关系方法
CN106959953A (zh) * 2016-01-08 2017-07-18 中国科学院声学研究所 一种基于统计特征的社交网络用户关系的计算方法

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10915919B2 (en) * 2017-01-03 2021-02-09 International Business Machines Corporation Topology-based online reward incentives

Also Published As

Publication number Publication date
CN107908673B (zh) 2019-06-25
CN107908673A (zh) 2018-04-13

Similar Documents

Publication Publication Date Title
WO2020253358A1 (zh) 业务数据的风控分析处理方法、装置和计算机设备
Adikari et al. Identifying fake profiles in linkedin
US11830004B2 (en) Blockchain transaction safety
JP6594329B2 (ja) 顔表現のためのシステムおよび方法
TWI505122B (zh) 用於自動管理安全性及/或隱私設定之方法、系統及電腦程式產品
US10936952B2 (en) Detecting content items in violation of an online system policy using templates based on semantic vectors representing content items
TW202022769A (zh) 風險辨識模型訓練方法、裝置及伺服器
CN108491714A (zh) 验证码的人机识别方法
Bahutair et al. Multi-perspective trust management framework for crowdsourced IoT services
WO2019051962A1 (zh) 社交平台用户的现实关系匹配方法、装置及可读存储介质
Lagraa et al. A simple graph embedding for anomaly detection in a stream of heterogeneous labeled graphs
WO2013160904A1 (en) Influence scores for social media profiles
CN110135978B (zh) 用户金融风险评估方法、装置、电子设备和可读介质
Liu et al. Keep your data locally: Federated-learning-based data privacy preservation in edge computing
Zhang et al. A trust model stemmed from the diffusion theory for opinion evaluation
WO2019080403A1 (zh) 社交平台用户的现实关系匹配方法、装置及可读存储介质
Kharaji et al. A new approach for finding cloned profiles in online social networks
CN109684454A (zh) 一种社交网络用户影响力计算方法及装置
CN114553456A (zh) 数字身份网络警报
Zhou et al. Cooperative malicious network behavior recognition algorithm in E-commerce
KR101928822B1 (ko) 사물 인터넷 환경에서 낯선 기기에 대한 사용자 신뢰도 계산 시스템 및 방법
US20230118341A1 (en) Inline validation of machine learning models
CN112597379B (zh) 数据识别方法、装置和存储介质及电子装置
WO2019080404A1 (zh) 跨社交平台用户匹配方法、数据处理装置及可读存储介质
Torabi et al. An ensemble classifier for link prediction in location based social network

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 24/09/2020)

122 Ep: pct application non-entry in european phase

Ref document number: 18869684

Country of ref document: EP

Kind code of ref document: A1