An internet information dissemination user role identification method and computer readable medium
By constructing an evolutionary cascade graph and combining weighted K-Shell decomposition and node propagation intensity information entropy method, user roles in Internet information propagation are identified, solving the problem that the role of secondary propagators has not been studied, and realizing the assessment of users' propagation capabilities and effective monitoring of information propagation.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- PEOPLE CN CO LTD
- Filing Date
- 2023-01-06
- Publication Date
- 2026-06-16
AI Technical Summary
In the current process of internet information dissemination, the role of secondary disseminators has not been fully studied, and existing K-Shell decomposition methods cannot accurately assess user influence in social media networks, making it difficult to effectively monitor and predict the information dissemination process.
By constructing an evolutionary cascade graph and combining weighted K-Shell decomposition and node propagation intensity information entropy method, we can identify initiators, guides, promoters and general users, and perform role analysis by utilizing the interaction characteristics between user nodes.
It enables accurate identification of user roles during the dissemination of internet information, assesses users' ability to spread information, and provides effective means of monitoring and predicting information dissemination.
Smart Images
Figure CN116228447B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the fields of data mining and network science, and in particular to a method for identifying user roles in Internet information dissemination and a computer-readable medium. Background Technology
[0002] With the advent of the information age, information devices and the internet have been widely used and popularized. The internet has gradually replaced traditional media, becoming the mainstream communication medium in the information age due to its efficiency. The popularization and development of the internet has greatly changed the channels through which people obtain data and information, and expanded the ways people communicate and interact. Online social networking platforms have become important channels for people to obtain and transmit content information, influencing people's activities in all aspects, including the economy, politics, sports, and entertainment. The impact of these platforms on people's activities mainly stems from two characteristics of the content: first, the authenticity of the disseminated content. Due to the virtual and unpredictable nature of online social platforms, many rumors and false information are mixed in with the content. When false content spreads in large quantities, it may mislead people's social and economic activities, and even trigger emotions such as panic and anger. Second, the real-time nature of the disseminated content. Social platforms contain a large number of users who share and disseminate a large amount of content in real time. Users can follow and obtain the latest hot events or content through channels such as trending topics. If we can analyze the roles of users participating in certain events by examining some potential public opinion or newer content on Weibo, it may help identify information manipulation and public opinion intervention, allowing governments and enterprises to prevent such incidents from happening in advance. Therefore, understanding the dissemination mechanism of content on social platforms, analyzing user roles in the process of internet information dissemination, and judging whether information manipulation exists in the dissemination process are of great significance for the prediction and regulation of information dissemination on social platforms.
[0003] Role analysis in internet information dissemination is generally categorized as a clustering problem. Existing research primarily employs traditional methods such as centrality, K-Shell decomposition, and information entropy-based approaches to identify important nodes in information dissemination networks. However, these methods focus on local and global characteristics to uncover influential nodes at the source of information dissemination, neglecting the role of secondary disseminators—non-source nodes—in information dissemination. These secondary disseminators are crucial to the information flow, thus requiring a method to measure whether they contribute to diffusion or inhibition. Furthermore, existing K-Shell decomposition methods treat edges in the same way when calculating the influence of disseminators in unweighted networks, applying only degree to link weights. However, in social media information dissemination networks, node degree does not accurately reflect user influence. Therefore, a novel link-weighted K-Shell decomposition method based on user interaction is needed for role analysis in social networks. Summary of the Invention
[0004] To address the aforementioned technical problems, this invention proposes a method for identifying user roles in internet information dissemination and a computer-readable medium.
[0005] The technical solution of this invention is a method for identifying user roles in internet information dissemination, specifically including the following steps:
[0006] Step 1: Construct an evolutionary cascade graph using each user node and each forwarded data entry from each user node.
[0007] Step 2: By traversing the source user nodes of the forwarded blog posts for each user node in the evolutionary cascade graph, multiple user nodes that published original blog posts are obtained, serving as multiple initiator user nodes;
[0008] Step 3: Sort the multiple forwarding data of each user node according to the posting time of the forwarded blog post, resulting in multiple sorted forwarding data for each user node. Using the first sorted forwarding data as the starting point, and combining it with earlier time intervals, select multiple sorted forwarding data as multiple early forwarding data for each user node. Search for the earliest forwarding time among the forwarding blog posts of these early forwarding data for each user node, and use this as the user node's forwarding time. Sort these forwarding times by time, resulting in multiple user nodes sorted by forwarding time. Using the forwarding time of the first user node as the starting point, and combining it with earlier time intervals, select multiple early intermediate user nodes. Finally, using a weighted K-Shell decomposition method, select multiple facilitator user nodes from the multiple forwarding data of each early intermediate user node.
[0009] Step 4: After sorting the user nodes by forwarding time, the forwarding time of the last early intermediate user node is used as the starting point. Combined with the peak time interval, multiple peak intermediate user nodes are selected. Combined with the multiple forwarding data of each peak intermediate user node, multiple promoter user nodes are obtained by filtering through the node propagation strength information entropy method.
[0010] Step 5: Divide the remaining user nodes (excluding the multiple initiator user nodes, multiple facilitator user nodes, and multiple promoter user nodes) into multiple general user nodes.
[0011] Step 6: Based on multiple initiator user nodes, multiple guide user nodes, multiple promoter user nodes, and multiple general user nodes, realize the positioning and analysis of users' network behavior roles in Internet information dissemination.
[0012] Preferably, each forwarded data from each user node in step 1 includes:
[0013] For each forwarded data entry of each user node, the forwarded blog post, the posting time of each forwarded blog post, the forwarding time of each forwarded blog post, the destination user node, the forwarding user node, the source user node, and the number of interaction operations with other user nodes in each forwarded data entry of each user node.
[0014] Preferably, step 3 involves combining multiple forwarding data from each early intermediate user node and filtering them using a weighted K-Shell decomposition method to obtain multiple mentor user nodes. The specific steps are as follows:
[0015] Step 3.1: Based on the destination user node of the forwarded blog post for each forwarded data of each early intermediate user node and the forwarding user node of the forwarded blog post for each forwarded data of each early intermediate user node, count the number of times each early intermediate user node forwards and is forwarded in multiple forwarded data, and use them as the in-degree and out-degree of the early intermediate user node respectively. The sum of the in-degree and out-degree of the early intermediate user node is used as the degree of the early intermediate user node.
[0016] Based on the number of interaction operations between each early intermediate user node and other user nodes in each forwarded data, the number of interaction operations between each early intermediate user node and other user nodes in multiple forwarded data is counted as the interaction intensity of the early intermediate user node.
[0017] Step 3.2: Calculate the link-weighted K-Shell value for each early intermediate user node. The specific calculation method is as follows:
[0018]
[0019] Where, d i Let R be the degree of the i-th early intermediate user node. i This represents the total number of neighboring nodes of the i-th early intermediate user node. This represents the j-th neighbor node of the i-th early intermediate user node, i.e., the... There are several early intermediate user nodes, where λ represents an adjustable parameter. This represents the propagation strength of the i-th early intermediate user node on its corresponding j-th neighbor node, i.e., the propagation strength of the i-th early intermediate user node on the j-th neighbor node. The propagation strength on early intermediate user nodes can be represented in a directed social network as the number of nodes i being affected by nodes i. Total number of reposts This represents the interaction strength of the i-th early intermediate user node with its corresponding j-th neighbor node, i.e., the interaction strength of the i-th early intermediate user node in the j-th node. The intensity of interaction on early intermediate user nodes;
[0020] Step 3.3: Remove the early intermediate user node with the smallest link weighted K-Shell value from multiple early intermediate user nodes, and add the removed early intermediate user node to the early intermediate user node set S-Shell;
[0021] Step 3.4: Repeat steps 3.1-3.3 until all early intermediate user nodes are added to the early intermediate user node set S-Shell;
[0022] Step 3.5: Sort all early intermediate user nodes in the S-Shell according to the link-weighted K-Shell value from largest to smallest, and select the top K early intermediate user nodes as multiple facilitator user nodes;
[0023] As a preferred method, step 4 involves combining multiple forwarding data points from intermediate user nodes during each peak period and filtering them using the node propagation strength information entropy method to obtain multiple promoter user nodes. The specific steps are as follows:
[0024] Step 4.1: Based on the destination user node of each forwarded blog post of each forwarded data of each user node and the forwarding user node of each forwarded blog post of each forwarded data of each user node, count the number of times the intermediate user node forwards multiple forwarded data in each peak period, and use it as the in-degree of the intermediate user node in that peak period.
[0025] Based on the destination user node of each forwarded blog post for each forwarded data of each user node, and the forwarding user node of each forwarded blog post for each forwarded data of each user node, the number of times the intermediate user node is forwarded in multiple forwarded data during each peak period is counted, which is taken as the out-degree of the intermediate user node during that peak period.
[0026] The information propagation intensity of intermediate user nodes during each peak period is calculated as follows:
[0027]
[0028] Where s(u) represents the information propagation intensity of intermediate user nodes during the u-th peak period. This represents the in-degree of the intermediate user node during the u-th peak period. This represents the out-degree of the intermediate user node during the u-th peak period;
[0029] Step 4.2: Construct a probability model for intermediate user nodes being selected by their neighboring nodes during peak periods, specifically defined as follows:
[0030]
[0031] Where P(u) represents the probability that the intermediate user node in the u-th peak period is selected by its neighboring nodes. This represents the information propagation strength of the v-th neighbor node of the intermediate user node during the u-th peak period, i.e., the... The information propagation intensity of intermediate user nodes during peak periods, F u This represents the total number of neighboring nodes of the intermediate user node during the u-th peak period;
[0032] Based on the probability model of a peak-period intermediate user node being selected by its neighboring nodes, the selection probability of the v-th neighboring node of the u-th peak-period intermediate user node is calculated and defined as:
[0033] Step 4.3: According to The propagation intensity information entropy of intermediate user nodes during the u-th peak period is calculated as follows:
[0034]
[0035] Among them, E u F represents the propagation intensity information entropy of intermediate user nodes during the u-th peak period. u This represents the total number of neighboring nodes of the intermediate user node during the u-th peak period;
[0036] Step 4.4: Select the top L nodes with the largest propagation intensity information entropy among the intermediate user nodes during the peak period to construct the information entropy user node set E-Shell, and select the top M nodes with the largest out-degree among the intermediate user nodes during the peak period to construct the out-degree user node set O-Shell. Take the intersection of E-Shell and O-Shell to obtain multiple promoter user nodes.
[0037] The present invention also provides a computer-readable medium storing a computer program executed by an electronic device, which, when run on the electronic device, causes the electronic device to perform the steps of the Internet information dissemination user role identification method.
[0038] The advantage of this invention is that it incorporates the interaction between user nodes as an important factor in the evaluation of Internet user dissemination capabilities, and also measures the information diffusion capabilities of user nodes in the dissemination process, ultimately using multiple features to achieve Internet dissemination user role analysis. Attached Figure Description
[0039] Figure 1 : Flowchart of the method according to an embodiment of the present invention. Detailed Implementation Plan
[0040] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0041] In specific implementation, the method proposed in the technical solution of this invention can be automatically executed by those skilled in the art using computer software technology. System devices for implementing the method, such as computer-readable storage media storing the corresponding computer program of the technical solution of this invention and computer equipment including the computer program running the corresponding computer program, should also be within the protection scope of this invention.
[0042] The following is combined with Figure 1 The technical solution of the method described in this invention is a method for user role recognition in Internet information dissemination, as detailed below:
[0043] Step 1: Construct an evolutionary cascade graph using each user node and each forwarded data entry from each user node.
[0044] Each forwarded data item for each user node mentioned in step 1 includes:
[0045] For each forwarded data entry of each user node, the forwarded blog post, the posting time of each forwarded blog post, the forwarding time of each forwarded blog post, the destination user node, the forwarding user node, the source user node, and the number of interaction operations with other user nodes in each forwarded data entry of each user node.
[0046] Step 2: By traversing the source user nodes of the forwarded blog posts for each user node in the evolutionary cascade graph, multiple user nodes that published original blog posts are obtained, serving as multiple initiator user nodes;
[0047] Step 3: Sort the multiple forwarding data of each user node according to the posting time of the forwarded blog post, resulting in multiple sorted forwarding data for each user node. Using the first sorted forwarding data as the starting point, and combining it with earlier time intervals, select multiple sorted forwarding data as multiple early forwarding data for each user node. Search for the earliest forwarding time among the forwarding blog posts of these early forwarding data for each user node, and use this as the user node's forwarding time. Sort these forwarding times by time, resulting in multiple user nodes sorted by forwarding time. Using the forwarding time of the first user node as the starting point, and combining it with earlier time intervals, select multiple early intermediate user nodes. Finally, using a weighted K-Shell decomposition method, select multiple facilitator user nodes from the multiple forwarding data of each early intermediate user node.
[0048] Step 3 involves combining multiple forwarding data points from each early intermediate user node and filtering them using a weighted K-Shell decomposition method to obtain multiple mentor user nodes. The specific steps are as follows:
[0049] Step 3.1: Based on the destination user node of the forwarded blog post for each forwarded data of each early intermediate user node and the forwarding user node of the forwarded blog post for each forwarded data of each early intermediate user node, count the number of times each early intermediate user node forwards and is forwarded in multiple forwarded data, and use them as the in-degree and out-degree of the early intermediate user node respectively. The sum of the in-degree and out-degree of the early intermediate user node is used as the degree of the early intermediate user node.
[0050] Based on the number of interaction operations between each early intermediate user node and other user nodes in each forwarded data, the number of interaction operations between each early intermediate user node and other user nodes in multiple forwarded data is counted as the interaction intensity of the early intermediate user node.
[0051] Step 3.2: Calculate the link-weighted K-Shell value for each early intermediate user node. The specific calculation method is as follows:
[0052]
[0053] Where, d i Let R be the degree of the i-th early intermediate user node. i This represents the total number of neighboring nodes of the i-th early intermediate user node. This represents the j-th neighbor node of the i-th early intermediate user node, i.e., the... An early intermediate user node, where λ = 0.5 represents an adjustable parameter. This represents the propagation strength of the i-th early intermediate user node on its corresponding j-th neighbor node, i.e., the propagation strength of the i-th early intermediate user node on the j-th neighbor node. The propagation strength on early intermediate user nodes can be represented in a directed social network as the number of nodes i being affected by nodes i. Total number of reposts This represents the interaction strength of the i-th early intermediate user node with its corresponding j-th neighbor node, i.e., the interaction strength of the i-th early intermediate user node in the j-th node. The intensity of interaction on early intermediate user nodes;
[0054] Step 3.3: Remove the early intermediate user node with the smallest link weighted K-Shell value from multiple early intermediate user nodes, and add the removed early intermediate user node to the early intermediate user node set S-Shell;
[0055] Step 3.4: Repeat steps 3.1-3.3 until all early intermediate user nodes are added to the early intermediate user node set S-Shell;
[0056] Step 3.5: Sort all early intermediate user nodes in the S-Shell according to the link-weighted K-Shell value from largest to smallest, and select the top K early intermediate user nodes as multiple facilitator user nodes;
[0057] Step 4: After sorting the user nodes by forwarding time, the forwarding time of the last early intermediate user node is used as the starting point. Combined with the peak time interval, multiple peak intermediate user nodes are selected. Combined with the multiple forwarding data of each peak intermediate user node, multiple promoter user nodes are obtained by filtering through the node propagation strength information entropy method.
[0058] Step 4 involves combining multiple forwarding data points from intermediate user nodes during each peak period and filtering them using the node propagation strength information entropy method to obtain multiple promoter user nodes. The specific steps are as follows:
[0059] Step 4.1: Based on the destination user node of each forwarded blog post of each forwarded data of each user node and the forwarding user node of each forwarded blog post of each forwarded data of each user node, count the number of times the intermediate user node forwards multiple forwarded data in each peak period, and use it as the in-degree of the intermediate user node in that peak period.
[0060] Based on the destination user node of each forwarded blog post for each forwarded data of each user node, and the forwarding user node of each forwarded blog post for each forwarded data of each user node, the number of times the intermediate user node is forwarded in multiple forwarded data during each peak period is counted, which is taken as the out-degree of the intermediate user node during that peak period.
[0061] The information propagation intensity of intermediate user nodes during each peak period is calculated as follows:
[0062]
[0063] Where s(u) represents the information propagation intensity of intermediate user nodes during the u-th peak period. This represents the in-degree of the intermediate user node during the u-th peak period. This represents the out-degree of the intermediate user node during the u-th peak period;
[0064] Step 4.2: Construct a probability model for intermediate user nodes being selected by their neighboring nodes during peak periods, specifically defined as follows:
[0065]
[0066] Where P(u) represents the probability that the intermediate user node in the u-th peak period is selected by its neighboring nodes. This represents the information propagation strength of the v-th neighbor node of the intermediate user node during the u-th peak period, i.e., the... The information propagation intensity of intermediate user nodes during peak periods, F u This represents the total number of neighboring nodes of the intermediate user node during the u-th peak period;
[0067] Based on the probability model of a peak-period intermediate user node being selected by its neighboring nodes, the selection probability of the v-th neighboring node of the u-th peak-period intermediate user node is calculated and defined as:
[0068] Step 4.3: According to The propagation intensity information entropy of intermediate user nodes during the u-th peak period is calculated as follows:
[0069]
[0070] Among them, E u F represents the propagation intensity information entropy of intermediate user nodes during the u-th peak period. u This represents the total number of neighboring nodes of the intermediate user node during the u-th peak period;
[0071] Step 4.4: Select the top L nodes with the largest propagation intensity information entropy among the intermediate user nodes during the peak period to construct the information entropy user node set E-Shell, and select the top M nodes with the largest out-degree among the intermediate user nodes during the peak period to construct the out-degree user node set O-Shell. Take the intersection of E-Shell and O-Shell to obtain multiple promoter user nodes.
[0072] Step 5: Divide the remaining user nodes (excluding the multiple initiator user nodes, multiple facilitator user nodes, and multiple promoter user nodes) into multiple general user nodes.
[0073] Step 6: Based on multiple initiator user nodes, multiple guide user nodes, multiple promoter user nodes, and multiple general user nodes, realize the positioning and analysis of users' network behavior roles in Internet information dissemination.
[0074] A specific embodiment of the present invention also provides a computer-readable medium.
[0075] The computer-readable medium is a server workstation;
[0076] The server workstation stores computer programs executed by the electronic device. When the computer program runs on the electronic device, it causes the electronic device to perform the steps of the Internet information dissemination user role recognition method according to the embodiments of the present invention.
[0077] It should be understood that any parts not described in detail in this specification belong to the prior art.
[0078] It should be understood that the above description of the preferred embodiments is quite detailed, but it should not be considered as a limitation on the scope of protection of this invention. Those skilled in the art, under the guidance of this invention, can make substitutions or modifications without departing from the scope of protection of the claims of this invention, and all such substitutions or modifications fall within the scope of protection of this invention. The scope of protection of this invention should be determined by the appended claims.
Claims
1. A method for identifying user roles in internet information dissemination, characterized in that, Includes the following steps: Step 1: Construct an evolutionary cascade graph using each user node and each forwarded data entry from each user node; Step 2: By traversing the source user nodes of the forwarded blog posts for each user node in the evolutionary cascade graph, multiple user nodes that published original blog posts are obtained, serving as multiple initiator user nodes; Step 3: Sort the reposted blog posts according to their posting time to obtain multiple early reposted data for each user node. Combine the multiple early reposted data of each user node and sort them by time to obtain multiple early intermediate user nodes. Combine the multiple reposted data of each early intermediate user node and filter them using the weighted K-Shell decomposition method to obtain multiple facilitator user nodes. Step 4: After sorting the user nodes by forwarding time, the forwarding time of the last early intermediate user node is used as the starting point. Combined with the peak time interval, multiple peak intermediate user nodes are selected. Combined with the multiple forwarding data of each peak intermediate user node, multiple promoter user nodes are obtained by filtering through the node propagation strength information entropy method. Step 5: Divide the remaining user nodes (excluding the multiple initiator user nodes, multiple facilitator user nodes, and multiple promoter user nodes) into multiple general user nodes. Step 6: Based on multiple initiator user nodes, multiple guide user nodes, multiple promoter user nodes, and multiple general user nodes, realize the positioning and analysis of users' network behavior roles in Internet information dissemination; Each forwarded data item for each user node mentioned in step 1 includes: The forwarded blog post for each forwarded data of each user node, the posting time of the forwarded blog post for each forwarded data of each user node, the forwarding time of the forwarded blog post for each forwarded data of each user node, the destination user node of the forwarded blog post for each forwarded data of each user node, the forwarding user node of the forwarded blog post for each forwarded data of each user node, the source user node of the forwarded blog post for each forwarded data of each user node, and the number of interaction operations with other user nodes in each forwarded data of each user node. Step 3 involves sorting the reposted posts according to their chronological order to obtain multiple early reposted posts for each user node, as detailed below: The multiple forwarded data of each user node are sorted according to the posting time of the forwarded blog post, resulting in multiple sorted forwarded data for each user node. The first sorted forwarded data in each user node is used as the starting point, and multiple sorted forwarded data are selected by combining the early time intervals, which are used as multiple early forwarded data for each user node.
2. The method for identifying user roles in internet information dissemination according to claim 1, characterized in that: Step 3 involves combining multiple early forwarding data from each user node and sorting them by time to obtain multiple early intermediate user nodes, as detailed below: The earliest forwarding time is searched among the forwarding times of multiple early forwarded blog posts for each user node as the forwarding time of the user node. The forwarding times of each user node are sorted by time to obtain multiple user nodes after sorting by forwarding time. Among the multiple user nodes after sorting by forwarding time, the forwarding time of the first user node is used as the time starting point. Multiple early intermediate user nodes are selected by combining the early time interval.
3. The method for identifying user roles in internet information dissemination according to claim 2, characterized in that: Step 3 involves combining multiple forwarding data points from each early intermediate user node and filtering them using a weighted K-Shell decomposition method to obtain multiple mentor user nodes. The specific steps are as follows: Step 3.1: Based on the destination user node of the forwarded blog post for each forwarded data of each early intermediate user node and the forwarding user node of the forwarded blog post for each forwarded data of each early intermediate user node, count the number of times each early intermediate user node forwards and is forwarded in multiple forwarded data, and use them as the in-degree and out-degree of the early intermediate user node respectively. The sum of the in-degree and out-degree of the early intermediate user node is used as the degree of the early intermediate user node. Based on the number of interaction operations between each early intermediate user node and other user nodes in each forwarded data, the number of interaction operations between each early intermediate user node and other user nodes in multiple forwarded data is counted as the interaction intensity of the early intermediate user node. Step 3.2: Calculate the link-weighted K-Shell value for each early intermediate user node. The specific calculation method is as follows: in, Let be the degree of the i-th early intermediate user node. This represents the total number of neighboring nodes of the i-th early intermediate user node. This represents the j-th neighbor node of the i-th early intermediate user node, i.e., the... There are several early intermediate user nodes, where λ represents an adjustable parameter. This represents the propagation strength of the i-th early intermediate user node on its corresponding j-th neighbor node, i.e., the propagation strength of the i-th early intermediate user node on the j-th neighbor node. The propagation strength on early intermediate user nodes can be represented in a directed social network as the number of nodes i being affected by nodes i. Total number of reposts This represents the interaction strength of the i-th early intermediate user node with its corresponding j-th neighbor node, i.e., the interaction strength of the i-th early intermediate user node in the j-th node. The intensity of interaction on early intermediate user nodes; Step 3.3: Remove the early intermediate user node with the smallest link weighted K-Shell value from multiple early intermediate user nodes, and add the removed early intermediate user node to the early intermediate user node set S-Shell; Step 3.4: Repeat steps 3.1-3.3 until all early intermediate user nodes are added to the early intermediate user node set S-Shell; Step 3.5: Sort all early intermediate user nodes in the S-Shell in descending order of their link-weighted K-Shell values, and select the top K early intermediate user nodes as multiple facilitator user nodes.
4. The method for identifying user roles in internet information dissemination according to claim 3, characterized in that: Step 4 involves combining multiple forwarding data points from intermediate user nodes during each peak period and filtering them using the node propagation strength information entropy method to obtain multiple promoter user nodes. The specific steps are as follows: Based on the destination user node of each forwarded blog post for each forwarded data of each user node, and the forwarding user node of each forwarded blog post for each forwarded data of each user node, the number of times the intermediate user node forwards multiple forwarded data in each peak period is counted, which is taken as the in-degree of the intermediate user node in that peak period. Based on the destination user node of each forwarded blog post for each forwarded data of each user node, and the forwarding user node of each forwarded blog post for each forwarded data of each user node, the number of times the intermediate user node is forwarded in multiple forwarded data during each peak period is counted, which is taken as the out-degree of the intermediate user node during that peak period. The information propagation intensity of intermediate user nodes during each peak period is calculated as follows: in, This represents the information propagation intensity of intermediate user nodes during the u-th peak period. This represents the in-degree of the intermediate user node during the u-th peak period. This represents the out-degree of the intermediate user node during the u-th peak period.
5. The method for identifying user roles in internet information dissemination according to claim 4, characterized in that: A probability model is constructed to determine whether an intermediate user node is selected by its neighboring nodes during peak periods. The specific definition is as follows: in, This represents the probability that a user node in the middle of the u-th peak period is selected by its neighboring nodes. This represents the information propagation strength of the v-th neighbor node of the intermediate user node during the u-th peak period, i.e., the... The intensity of information propagation among intermediate user nodes during peak periods This represents the total number of neighboring nodes of the intermediate user node during the u-th peak period; Based on the probability model of a peak-period intermediate user node being selected by its neighboring nodes, the selection probability of the v-th neighboring node of the u-th peak-period intermediate user node is calculated and defined as: ; according to The propagation intensity information entropy of the intermediate user node in the u-th peak period is calculated as follows: in, This represents the propagation intensity information entropy of the intermediate user nodes during the u-th peak period. This represents the total number of neighboring nodes of the intermediate user node during the u-th peak period.
6. The method for identifying user roles in internet information dissemination according to claim 5, characterized in that: The L nodes with the highest propagation intensity information entropy among the intermediate user nodes during the peak period are selected to construct the information entropy user node set E-Shell. The M nodes with the highest out-degree among the intermediate user nodes during the peak period are selected to construct the out-degree user node set O-Shell. The intersection of E-Shell and O-Shell is taken to obtain multiple promoter user nodes.
7. A computer-readable medium, characterized in that, It stores a computer program executed by an electronic device, which, when run on the electronic device, causes the electronic device to perform the steps of the method as described in any one of claims 1-6.