Method for determining importance of users in social network
By converting social networks into graph structures and using α-random walk algorithms, the problem of inefficient measurement of user importance in social networks in the prior art is solved, and efficient Page ranking score estimation in large-scale networks is achieved and user importance assessment.
Patent Information
- Application Number
- CN202411167723.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-23
- Publication Date
- 2024-12-03
AI Technical Summary
The measurement of user importance in existing social networks is not efficient in calculating the estimation of Page ranking scores, and it is difficult to support the structure of super-large-scale social networks.
The relationship between users and users in social networks is converted into a graph structure, and the α-random walk algorithm is used to calculate the stop probability starting from the target node, and the Page ranking score of the target node is estimated, thereby determining the importance of the user.
Efficiently complete the estimation of Page's ranking scores on large-scale social networks to ensure that the relative error between the approximate calculation results and the real results meets the actual requirements, and simplifies the user's importance evaluation process.
Smart Images

Figure CN119066274A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of computer technology, and in particular to a method for determining the importance of a user in a social network. Background Art
[0002] With the rapid development of my country's Internet industry, social networks represented by WeChat and Weibo have become one of the most widely used Internet applications. Relying on Internet technology, they provide a platform for communication and interaction for people's social activities through modern communication tools such as computers and mobile phones.
[0003] Measuring the importance of social users on social networks is one of the issues that social software operators, advertising companies, government agencies, etc. are highly concerned about. For example, advertising companies will focus on highly active users on social networks (such as Weibo big V users) and improve product promotion efficiency by launching targeted business invitations to them; government regulatory agencies also need to monitor key users in real time to maintain network security and prevent the spread of rumors.
[0004] The existing mainstream method for measuring the importance of social users is to calculate the estimated Page Rank score of social users on social networks.
[0005] However, existing methods for estimating the Page Rank scores of social users are not efficient and are difficult to support ultra-large-scale social network structures.
[0006] The information disclosed in this background technology section is only intended to enhance the understanding of the overall background of the invention and should not be regarded as an acknowledgment or any form of suggestion that the information constitutes the prior art already known to a person skilled in the art. Summary of the invention
[0007] The purpose of the present invention is to provide a method for determining the importance of a user in a social network, which can solve the problem that the existing method for estimating the Page Rank score of a social user is inefficient in calculation.
[0008] To achieve the above object, the present invention provides a method for determining the importance of a user in a social network, comprising:
[0009] Converting users and relationships between users in a social network into a graph structure, wherein the graph structure includes n nodes corresponding to the users, m edges corresponding to the relationships between the users, and the target user corresponds to the target node t in the graph structure;
[0010] For any node s on the graph structure, calculate the estimated probability of an α-random walk starting from the target node t stopping at node s, where α is the random walk attenuation factor, and an α-random walk is a walk that stops at node s with a probability of α and moves to a randomly selected neighbor node v of node s with a probability of 1-α;
[0011] Update node s to another node among n nodes in the graph structure;
[0012] Repeat the step of calculating the estimated probability of the α-random walk starting from the target node t stopping at the node s until the estimated probability of the α-random walk starting from the target node t stopping at all nodes s in the graph structure is obtained;
[0013] Based on the estimated probability of an α-random walk starting from the target node t stopping at all nodes s in the graph structure, the estimated Page rank score of the target node t is obtained
[0014] According to the estimation of the Page ranking score of the target node t The importance of the target user corresponding to the target node t is determined.
[0015] In one embodiment, the probability of the α-random walk starting from the target node t stopping at all nodes s in the graph structure is estimated to obtain an estimate of the Page Rank score of the target node t. include:
[0016] According to formula 1 The formula 1 is:
[0017]
[0018] Among them, d t is the number of neighbors of the target node t on the graph structure, d s is the number of neighbors of node s in the graph structure, is the estimate of the probability that an α-random walk starting from the target node t stops at the node s, and V is the set of all nodes in the graph.
[0019] In one embodiment, for any node s on the graph structure, calculating an estimate of the probability that an α-random walk starting from a target node t stops at the node s includes:
[0020] Step S21: for any node s on the graph structure, estimate the probability that the α-random walk starting from the target node t stops at the node s. The initial value of is set to 0;
[0021] Step S22: Generate an α-random walk from the target node t. If the α-random walk stops at the node s, update according to formula (2): Values:
[0022]
[0023] Step S23, repeat step S22n r When the nth r When step S22 is executed for the second time, the updated As an estimate of the probability that an α-random walk starting from the target node t stops at the node s , where n r is the default value.
[0024] In one embodiment, generating an α-random walk from the target node t in step S22 includes:
[0025] Step S221, taking the target node t as the current node of the α-random walk, the α-random walk executes step S222 with a probability of α, and executes step S223 with a probability of 1-α;
[0026] Step S222, the α-random walk ends, and the current node of the α-random walk is returned as the stop node of the α-random walk;
[0027] Step S223, uniformly and randomly select a node u from all neighboring nodes of the current node of the α-random walk, and use node u as the current node of the α-random walk; the α-random walk ends with a probability of α, and returns the current node of the α-random walk as the stop node of the α-random walk; repeat step S223 with a probability of 1-α until the random walk ends.
[0028] In one embodiment, the α-random walk stops at a node s including:
[0029] Node s is the stop node of the α-random walk.
[0030] In one embodiment, the n r Determine according to formula (3):
[0031]
[0032] Among them, α is the random walk attenuation factor, α∈(0,1), p f is the failure probability parameter, p f ∈(0,1), c∈(0,1) is the relative error parameter, d min Represents the minimum number of neighbors of a node on the graph structure G, d tis the number of neighbors of the target node t on the graph structure, It is the square root of the number of edges m in the graph structure.
[0033] Compared with the prior art, the method for determining the importance of users in a social network provided by this embodiment can efficiently complete the calculation of the estimated Page Rank score of a target social user on a large-scale social network, while ensuring that the relative error between the approximate calculation result and the true result meets the actual requirements, and then determine the importance of the user in the social network based on the estimated Page Rank score of the user. BRIEF DESCRIPTION OF THE DRAWINGS
[0034] Figure 1 is a flow chart of a method for determining user importance in a social network according to an embodiment of the present invention. DETAILED DESCRIPTION
[0035] The specific implementation modes of the present invention are described in detail below in conjunction with the accompanying drawings, but it should be understood that the protection scope of the present invention is not limited by the specific implementation modes.
[0036] Unless explicitly stated otherwise, throughout the specification and claims, the term “comprise” or variations such as “include” or “comprising”, etc., will be understood to include the stated elements or components but not to exclude other elements or components.
[0037] At present, the existing mainstream method for measuring the importance of social users is to calculate the estimated Page Rank score of social users on social networks. The Page Rank score of social users is defined based on the following idea: there may be friendships between social users. If a social user has many friends, or a social user is a friend of another known important social user, then the social user can be considered to have a high importance.
[0038] Therefore, by calculating the estimated Page Rank score of a user in the social network, the importance of the user in the social network can be determined.
[0039] like Figure 1 As shown, it is a flow chart of a method for determining user importance in a social network according to a preferred embodiment of the present invention, including steps S1-S6.
[0040] Step S1, converting users and relationships between users in a social network into a graph structure G, wherein the graph structure includes n nodes corresponding to the users, m edges corresponding to the relationships between the users, and the target user corresponds to the target node t in the graph structure.
[0041] It should be noted that in step S1, the users and the relationships between users in the social network are converted into a graph structure G, which includes nodes corresponding to the users and edges corresponding to the relationships between the users. In particular, the embodiment is dedicated to calculating the estimated Page Rank score of the target user in the social network, and the target user is converted into a node t on the graph structure G, and node t is called the target node. The graph structure G includes n nodes and m edges. The estimated Page Rank score of the target node t on the graph structure G can reflect the importance of the target user on the social network. The larger the estimated Page Rank score of the target node t, the higher the importance of the target user on the social network.
[0042] Specifically, if there is a friend relationship between social user u and social user v, an undirected edge is established from node u to node v on the graph structure G. We call all nodes that have an edge relationship with node u the neighbor nodes of node u; the set of all neighbor nodes of node u is called the neighbor set of node u; the number of nodes that have an edge relationship with node u on the graph structure is called the number of neighbors of node u, also called the degree of node u, denoted as d u In addition, using d min Represents the minimum value of the node degree on the graph structure G. V represents the set of all nodes on the graph.
[0043] Step S2, for any node s on the graph structure, calculate the estimated probability of an α-random walk starting from the target node t stopping at node s, where α is the random walk attenuation factor, and the α-random walk is a walk that stops at node s with a probability of α and moves to a randomly selected neighbor node v of node s with a probability of 1-α.
[0044] In one embodiment, α∈(0,1), the specific value of α is set by the user of this embodiment according to actual conditions, and the conventional setting of α is 0.2.
[0045] Step S3, updating the node s to another node of the n nodes in the graph structure;
[0046] Step S4, repeatedly calculating the estimated probability of the α-random walk starting from the target node t stopping at the node s, until the estimated probability of the α-random walk starting from the target node t stopping at all nodes s in the graph structure is obtained;
[0047] Step S5, based on the estimated probability of the α-random walk starting from the target node t stopping at all nodes s in the graph structure, obtain the estimated Page Rank score of the target node t
[0048] Step S6, based on the estimated PageRank score of the target node t The importance of the target user corresponding to the target node t is determined.
[0049] Therefore, through the method for determining the importance of users in a social network provided by this embodiment, the estimated calculation of the Page Rank score of a target social user can be efficiently completed on a large-scale social network, while ensuring that the relative error between the approximate calculation result and the true result meets the actual requirements.
[0050] In one embodiment, step S2 may specifically include:
[0051] Step S21: for any node s on the graph structure, estimate the probability that the α-random walk starting from the target node t stops at the node s. The initial value of is set to 0;
[0052] Step S22: Generate an α-random walk from the target node t. If the α-random walk stops at the node s, update according to formula (2): Values:
[0053]
[0054] Step S23, repeat step S22n r When the nth r When step S22 is executed for the second time, that is, n is generated from node t r When the α-random walk is performed, the updated As an estimate of the probability that an α-random walk starting from the target node t stops at the node s, where n r is the default value.
[0055] In one embodiment, generating an α-random walk from the target node t in step S22 may specifically include: steps S221 to S223.
[0056] Step S221, taking the target node t as the current node of the α-random walk, the α-random walk executes step S222 with a probability of α, and executes step S223 with a probability of 1-α;
[0057] Step S222, the α-random walk ends, and the current node of the α-random walk is returned as the stop node of the α-random walk;
[0058] Step S223, uniformly and randomly select a node u from all neighboring nodes of the current node of the α-random walk, and use node u as the current node of the α-random walk; the α-random walk ends with a probability of α, and returns the current node of the α-random walk as the stop node of the α-random walk; repeat step S223 with a probability of 1-α until the random walk ends.
[0059] In one embodiment, the α-random walk in step S22 stops at node s, which may include: node s is the stop node of the α-random walk. In other words, if the stop node returned by the α-random walk generated from the target node t is node s, then update according to formula (2): The value of .
[0060] In one embodiment, S5 includes: calculating according to formula 1 The formula 1 is:
[0061]
[0062] Among them, d t is the number of nodes that have an edge relationship with the target node t in the graph structure, also called the number of neighbors of the target node t; d s It is the number of nodes that have an edge relationship with node s in the graph structure, also called the number of neighbors of node s. is an estimate of the probability that an α-random walk starting from the target node t stops at the node s.
[0063] In one embodiment, the n r It can be determined according to formula (3):
[0064]
[0065] Among them, α is the random walk attenuation factor, α∈(0,1), p f is the failure probability parameter, p f ∈(0,1), c∈(0,1) is the relative error parameter, d min Represents the minimum number of neighbors of a node on the graph structure G, d t is the number of neighbors of the target node t on the graph structure, It is the square root of the total number of edges m on the graph structure G. f and c are specified by the user of this embodiment according to actual conditions. For example, p f is 0.1, c is 0.1.
[0066] Specifically, here n r It is a value set by the user of this embodiment according to actual conditions and is related to the Page ranking score π(t).
[0067] The calculation process of Page Ranking Score π(t) can refer to the following steps:
[0068] The Page ranking score π(t) of the target node t can be calculated by formula (4):
[0069]
[0070] Among them, π(s, t) represents the probability that an α-random walk starting from a graph node s on the graph G stops at node t, V represents the set of all nodes on the graph G, and n represents the number of nodes on the graph G.
[0071] The existing mainstream social user importance measurement methods hope to calculate the estimated PageRank score of social users on social networks, that is, to calculate the estimated PageRank score of the target node t on the graph G. And guarantee The relative error between the Page ranking score π(t) of the target node t on the graph G does not exceed c, that is, the probability that the following formula is true is not less than 1-p f :
[0072]
[0073] Then we need to set:
[0074]
[0075] Thus we get formula (3). Here, p f ∈(0,1) is the failure probability parameter, and c∈(0,1) is the relative error parameter. f and c are values set by the user of this embodiment according to actual conditions. min Represents the minimum value of the node degree on the graph structure G.
[0076] The method of this embodiment can Get the estimated value of the Page ranking score of the target node t within the time And it can be guaranteed that the estimated value of the Page ranking score of the target node t calculated by this method is The probability that the relative error between the value and the exact value π(t) does not exceed c is no less than 1-p f ,Right now: The probability of establishment is not less than 1-p f In addition, the steps of the method proposed in this embodiment are simpler than most previous methods, which is conducive to the specific implementation and application of the method.
[0077] In contrast, the optimal algorithm before this method can only It takes time to get the estimated Page ranking score of the target node t And guarantee the estimated value The probability that the relative error between the value and the exact value π(t) does not exceed c is no less than 1-p f , and the method steps are very complicated.
[0078] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.
[0079] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0080] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0081] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0082] The foregoing description of specific exemplary embodiments of the present invention is for the purpose of illustration and demonstration. These descriptions are not intended to limit the present invention to the precise form disclosed, and it is clear that many changes and variations can be made based on the above teachings. The purpose of selecting and describing the exemplary embodiments is to explain the specific principles of the present invention and its practical application, so that those skilled in the art can realize and utilize various different exemplary embodiments of the present invention and various different selections and changes. The scope of the present invention is intended to be limited by the claims and their equivalents.
Claims
1. A method for determining the importance of a user in a social network, characterized in that: include: Converting users and relationships between users in a social network into a graph structure, wherein the graph structure includes n nodes corresponding to the users, m edges corresponding to the relationships between the users, and the target user corresponds to a target node t in the graph structure; For any node s on the graph structure, calculate the estimated probability of an α-random walk starting from the target node t stopping at node s, where α is the random walk attenuation factor, and an α-random walk is a walk that stops at node s with a probability of α and moves to a randomly selected neighbor node v of node s with a probability of 1-α; Update node s to another node among n nodes in the graph structure; Repeat the step of calculating the estimated probability of the α-random walk starting from the target node t stopping at the node s until the estimated probability of the α-random walk starting from the target node t stopping at all nodes s in the graph structure is obtained; Based on the estimated probability of an α-random walk starting from the target node t stopping at all nodes s in the graph structure, the estimated Page rank score of the target node t is obtained According to the estimation of the Page ranking score of the target node t The importance of the target user corresponding to the target node t is determined.
2. The determination method according to claim 1, characterized in that: The estimated probability of the α-random walk starting from the target node t stopping at all nodes s in the graph structure, and the estimated Page rank score of the target node t is obtained include: According to formula 1 The formula 1 is: Among them, d t is the number of neighbors of the target node t on the graph structure, d s is the number of neighbors of node s in the graph structure, is the estimate of the probability that an α-random walk starting from the target node t stops at the node s, and V is the set of all nodes in the graph.
3. The determination method according to claim 1, characterized in that: For any node s on the graph structure, the probability of an α-random walk starting from the target node t stopping at node s is estimated, including: Step S21: for any node s on the graph structure, estimate the probability that the α-random walk starting from the target node t stops at the node s. The initial value of is set to 0; Step S22, generate an α-random walk from the target node t. If the α-random walk stops at the node s, update according to formula (2): Values: Step S23, repeat step S22 n r When the nth r When step S22 is executed for the second time, the updated As an estimate of the probability that an α-random walk starting from the target node t stops at the node s Among them, n r is the default value.
4. The determination method according to claim 3, characterized in that: The step S22 generates an α-random walk from the target node t, including: Step S221, taking the target node t as the current node of the α-random walk, the α-random walk executes step S222 with a probability of α, and executes step S223 with a probability of 1-α; Step S222, the α-random walk ends, and the current node of the α-random walk is returned as the stop node of the α-random walk; Step S223, uniformly and randomly select a node u from all neighboring nodes of the current node of the α-random walk, and use node u as the current node of the α-random walk; the α-random walk ends with a probability of α, and returns the current node of the α-random walk as the stop node of the α-random walk; repeat step S223 with a probability of 1-α until the random walk ends.
5. The determination method according to claim 3, characterized in that: α-Random walk stops at node s including: Node s is the stop node of the α-random walk.
6. The determination method according to claim 3, characterized in that: The r Determine according to formula (3): Among them, α is the random walk attenuation factor, α∈(0,1), p f is the failure probability parameter, p f ∈(0,1), c∈(0,1) is the relative error parameter, d min Represents the minimum number of neighbors of a node on the graph structure G, d t is the number of neighbors of the target node t on the graph structure, It is the square root of the number of edges m in the graph structure.
