A risk transmission path deduction and intervention method and system based on a parallel simulation environment

Abstract

The application provides a risk transmission path deduction and intervention method and system based on a parallel simulation environment, acquires social network interaction data containing user attribute characteristics and historical information propagation records; analyzes the historical information propagation records and the user attribute characteristics, generates personalized information propagation parameters and configures the personalized information propagation parameters to virtual user behavior entities corresponding to preset real social network users, generates a completed virtual user behavior entity set; drives the completed virtual user behavior entity set to simulate information propagation in a parallel simulation environment constructed based on the social network interaction data, generates an information flow path; analyzes the information flow path to evaluate the influence degree of a plurality of information intervention strategies on the completed virtual user behavior entity set in the parallel simulation environment, and generates a final intervention scheme. The application realizes accurate deduction of a risk transmission path in a virtual environment and verification of the effectiveness of an intervention strategy, and reduces the risk and cost brought by direct experiments in a real network.

Description

Technical Field

[0001] This invention relates to the field of path deduction and intervention technology, and in particular to a method and system for risk transmission path deduction and intervention based on a parallel simulation environment. Background Technology

[0002] As social networks play an increasingly prominent role in information dissemination, the rapid spread of risk events such as misinformation and public opinion crises poses a serious challenge to social stability and public safety, urgently requiring a technological means to accurately depict the dynamic evolution of information among complex user groups. Against this backdrop, predicting potential spread paths before risk events occur and assessing the actual effectiveness of different intervention measures has become a key requirement for current information dissemination risk management. This scenario demands that the technological solution not only reproduce the differences in user behavior and interaction structures in real social networks but also possess the ability to dynamically simulate multiple rounds of information transmission. This would provide decision-makers with quantifiable path projection results and intervention strategy support, enabling a shift from passive response to proactive prevention and control.

[0003] The current mainstream approach for risk path prediction combines multi-agent modeling with historical propagation graph reconstruction. This approach mines public propagation chain data from social media platforms to construct a temporal dependency graph of information diffusion. It then trains an individual forwarding probability prediction model using user profile features. The prediction results are subsequently embedded into a multi-agent system, driving agents to simulate information transmission according to statistical laws. Finally, through graph backtracking, it identifies potential high-influence node combinations, forming a set of potential risk propagation paths. This method achieves a degree of automated deduction, avoiding the subjective limitations of relying solely on expert experience, and has already been preliminarily applied in some public opinion monitoring systems. Existing solutions have some inherent flaws, including relying on static historical propagation patterns for modeling, which makes it difficult to fully reflect the adaptive adjustments of user behavior as the environment changes, resulting in a lack of dynamic feedback mechanisms for individual decision-making logic during the simulation process; when constructing agent behavior rules, group average parameters are usually used, which fails to effectively integrate fine-grained individual attributes and interaction context information, resulting in insufficient generalization ability of simulation results when facing sudden and atypical propagation events; and the lack of a closed-loop simulation architecture that is continuously aligned with the real network means that the intervention suggestions generated cannot be validated under conditions close to the real propagation environment, thereby reducing the actual control effectiveness after policy deployment. Summary of the Invention

[0004] This invention provides a method and system for risk transmission path deduction and intervention based on a parallel simulation environment. It addresses several issues in existing technologies, such as reliance on static historical propagation patterns for modeling, which fails to adequately reflect the adaptive adjustments of user behavior to environmental changes, resulting in a lack of dynamic feedback mechanisms for individual decision-making logic during simulation; the use of group average parameters in constructing agent behavior rules, which fails to effectively integrate fine-grained individual attributes and interaction context information, leading to insufficient generalization ability of simulation results in the face of sudden and atypical propagation events; and the lack of a closed-loop simulation architecture continuously aligned with the real network, preventing the effectiveness verification of generated intervention suggestions under near-real propagation environment conditions, thus reducing the actual regulatory efficiency after policy deployment.

[0005] In a first aspect, the present invention provides a method for risk transmission path deduction and intervention based on a parallel simulation environment, comprising: Acquire social network interaction data, wherein the social network interaction data includes user attribute characteristics and historical information dissemination records; Statistical analysis and processing are performed on the historical information dissemination records and the user attribute characteristics to generate personalized information dissemination parameters; The personalized information dissemination parameters are configured in virtual user behavior entities corresponding to users of a preset real social network to generate a set of configured virtual user behavior entities. In a parallel simulation environment constructed based on the social network interaction data, the configured set of virtual user behavior entities is driven to simulate information propagation and generate information flow paths. The information flow bottlenecks in the information flow path are analyzed, and the key transmission nodes connected to different virtual user behavior entities in the analysis results are connected to generate risk transmission paths. Based on the risk transmission path, the impact of various information intervention strategies in the parallel simulation environment on the configured set of virtual user behavior entities is evaluated. The information intervention strategy whose impact meets the preset optimal impact condition is selected as the final intervention plan, and the final intervention plan is deployed in the preset real social network.

[0006] Optionally, social network interaction data is acquired, wherein the social network interaction data includes user attribute characteristics and historical information dissemination records, including: Utilize preset data access points to receive raw interactive data streams from preset real social network data sources; The user basic attribute data in the original interactive data stream is subjected to structured transformation processing to generate user attribute features; Based on the time series of the original interactive data stream, the discrete propagation events in the original interactive data stream are recombined to generate historical information propagation records; The user attribute features and the historical information dissemination records are linked and integrated to generate social network interaction data.

[0007] Optionally, statistical analysis is performed on the historical information dissemination records and the user attribute characteristics to generate personalized information dissemination parameters, including: The propagation pattern features are extracted from the historical information propagation records, and the propagation pattern features are statistically analyzed to generate the user's historical propagation pattern. The user's social influence index is generated by performing a weighted calculation on the number of followers and interaction frequency data in the user attribute features. The user's historical propagation pattern and the user's social influence index are fused and calculated to generate a propagation tendency parameter; Extract the information forwarding probability and content preference weight from the propagation tendency parameters; The information forwarding probability and the content preference weight are configured into each virtual user behavior entity corresponding to users of a preset real social network to generate personalized information dissemination parameters.

[0008] Optionally, the personalized information dissemination parameters are configured in virtual user behavior entities corresponding to users of a preset real social network to generate a configured set of virtual user behavior entities, including: Create virtual user behavior entities that correspond one-to-one with users in each preset real social network; The virtual user behavior entities are combined to generate an initial set of virtual user behavior entities; Establish a correspondence between each of the virtual user behavior entities and each of the preset real social networks; Based on the correspondence, the information forwarding probability in the personalized information dissemination parameters is allocated to each of the virtual user behavior entities to generate a probability configuration result; Based on the correspondence, the content preference weights in the personalized information dissemination parameters are allocated to each of the virtual user behavior entities to generate weight configuration results; Based on the probability configuration result and the weight configuration result, the initial virtual user behavior entity set is updated to generate a configured virtual user behavior entity set.

[0009] Optionally, in a parallel simulation environment constructed based on the social network interaction data, the configured set of virtual user behavior entities is driven to simulate information propagation and generate information flow paths, including: A parallel simulation environment is constructed using the user relationship topology in the aforementioned social network interaction data; The configured set of virtual user behavior entities is loaded into the nodes of the parallel simulation environment to generate node loading results; The preset initial propagation information is input into the parallel simulation environment to trigger the virtual user behavior entities configured in the node loading result to perform simulated interaction behavior and generate simulated interaction behavior data. The transmission direction of the preset initial propagation information is determined based on each configured virtual user behavior entity to obtain information transmission direction data; Based on the information transmission direction data, multiple initial information propagation paths are extracted from the information transmission trajectory of the simulated interaction behavior data; The topological structures in each of the initial information propagation paths are integrated to generate an information flow path.

[0010] Optionally, the information flow bottlenecks in the information flow path are analyzed, and the key transmission nodes connected to different virtual user behavior entities in the analysis results are connected to generate a risk transmission path, including: Information transmission delay points in the information flow path are detected to identify the bottleneck locations of information flow from the detection results; Analyze the node connection relationships at the bottleneck locations of the information flow to determine the hub nodes that connect different virtual user behavior entities; Each key transmission node in the hub node is marked to generate a key node set; Based on the set of key nodes, establish the connection relationship between each key transmission node, traverse each connection relationship, and generate multiple initial risk transmission paths; Each initial risk transmission path is verified for path integrity, and the initial risk transmission path corresponding to the verification result that meets the preset pass conditions is taken as the risk transmission path.

[0011] Optionally, based on the risk transmission path, the impact of various information intervention strategies in the parallel simulation environment on the configured set of virtual user behavior entities is evaluated, and the information intervention strategy whose impact meets the preset optimal impact condition is selected as the final intervention plan. The final intervention plan is then deployed to the preset real social network, including: At each of the key transmission nodes along the risk transmission path, various information intervention strategies in the parallel simulation environment are executed to generate strategy application results; The behavioral change characteristics of the set of virtual user behavior entities configured in the application results of the strategy are monitored in order to collect behavioral change data from the monitoring results; The behavioral change data are analyzed, and the impact index of each information intervention strategy is calculated based on the analysis results. The various influence degree indicators are compared to select the influence degree indicator that meets the preset optimal influence degree condition as the target influence degree indicator. Select the information intervention strategy corresponding to the target impact level index as the final intervention plan; The final intervention plan is mapped onto the preset real social network to complete the deduction and intervention of the risk transmission path.

[0012] Secondly, the present invention provides a risk transmission path simulation and intervention system based on a parallel simulation environment, comprising: The acquisition module is used to acquire social network interaction data, wherein the social network interaction data includes user attribute characteristics and historical information dissemination records; The processing module is used to perform statistical analysis and processing on the historical information dissemination records and the user attribute characteristics to generate personalized information dissemination parameters; The configuration module is used to configure the personalized information dissemination parameters into virtual user behavior entities corresponding to users of a preset real social network, so as to generate a set of configured virtual user behavior entities. The driving module is used to drive the configured set of virtual user behavior entities to simulate information propagation and generate information flow paths in a parallel simulation environment constructed based on the social network interaction data. The connection module is used to analyze the information flow bottlenecks in the information flow path, connect the key transmission nodes connected to different virtual user behavior entities in the analysis results, and generate a risk transmission path. The evaluation module is used to evaluate the impact of various information intervention strategies in the parallel simulation environment on the configured set of virtual user behavior entities based on the risk transmission path, select the information intervention strategy whose impact meets the preset optimal impact condition as the final intervention plan, and deploy the final intervention plan into the preset real social network.

[0013] Thirdly, the present invention provides a computing device, including a processor and a memory, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute a risk transmission path deduction and intervention method based on a parallel simulation environment as described in any of the first aspects.

[0014] Fourthly, the present invention provides a computer storage medium storing computer program instructions thereon, wherein the computer program instructions, when executed by a processor, implement the risk transmission path deduction and intervention method based on a parallel simulation environment as described in any one of the first aspects.

[0015] This invention acquires social network interaction data and generates personalized information propagation parameters based on historical information dissemination records and user attribute characteristics. It constructs a set of virtual user behavior entities corresponding to real social networks, drives the entity set to simulate information propagation in a parallel simulation environment to generate information flow paths, analyzes bottlenecks in the paths and connects key transmission nodes to form risk transmission paths, and finally selects the optimal solution to deploy on real social networks by evaluating the impact of various intervention strategies. This enables accurate simulation of risk transmission paths and verification of the effectiveness of intervention strategies in a virtual environment, greatly reducing the risks and costs of direct experimentation in real networks.

[0016] Furthermore, by creating virtual user behavior entities that correspond one-to-one with real social network users and establishing mapping relationships, the information forwarding probability and content preference weights in the personalized information dissemination parameters are accurately allocated to the corresponding virtual entities, generating a configured entity set. This ensures that the dissemination characteristics of each behavior entity in the virtual simulation environment are highly consistent with real user behavior, laying a solid foundation for the accurate deduction of subsequent risk transmission paths and the effective verification of intervention strategies.

[0017] These or other aspects of the invention will become more apparent from the following description of the embodiments. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A flowchart of a risk transmission path deduction and intervention method based on a parallel simulation environment is provided for an embodiment of the present invention; Figure 2 A schematic diagram of a risk transmission path deduction and intervention system based on a parallel simulation environment is provided for an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of a computing device provided in an embodiment of the present invention. Detailed Implementation

[0020] To enable those skilled in the art to better understand the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0021] In some of the processes described in the specification, claims, and accompanying drawings of this invention, multiple operations appearing in a specific order are included. However, it should be clearly understood that these operations may not be executed in the order they appear herein, or may be executed in parallel. The operation numbers, such as 101, 102, etc., are merely used to distinguish different operations and do not represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the descriptions such as "first," "second," etc., in this document are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to different types.

[0022] 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.

[0023] Figure 1 A flowchart of a risk transmission path deduction and intervention method based on a parallel simulation environment is provided for an embodiment of the present invention, as shown below. Figure 1 As shown, the method includes: Existing risk intervention schemes based on social networks suffer from three key drawbacks: First, traditional analysis methods struggle to dynamically simulate the complex interactive behaviors of large-scale user groups, failing to accurately reconstruct the true path of information dissemination. Second, static network analysis models cannot capture real-time bottlenecks and key transmission nodes in the information flow process, resulting in insufficient risk prediction accuracy. Third, intervention strategies are typically implemented directly in real social networks, lacking pre-implementation verification mechanisms, potentially triggering uncontrollable chain reactions. To address these issues, this invention proposes a method for risk transmission path deduction and intervention based on a parallel simulation environment. This involves constructing a simulation environment parallel to real social networks, transforming real user attributes and historical behavioral data into personalized parameters for virtual entities, simulating the entire information dissemination process within the simulation environment, accurately identifying key nodes and bottlenecks in the risk transmission path, and conducting simulation tests and effect evaluations of various intervention strategies. Finally, the validated optimal strategy is securely deployed to the real network environment, thereby achieving pre-emptive prediction and controllable intervention of risk transmission. Based on this, this invention provides a method for risk transmission path deduction and intervention in a parallel simulation environment, such as... Figure 1 ,include: Step 101: Obtain social network interaction data, wherein the social network interaction data includes user attribute characteristics and historical information dissemination records.

[0024] In this step, social network interaction data refers to a multidimensional dataset obtained from social platforms that includes user relationship networks, attribute data, and behavioral records; user attribute characteristics refer to static attribute data including user age, gender, region, and interest tags; and historical information dissemination records refer to time-series data reflecting users' historical posting, forwarding, commenting, and other dissemination behaviors.

[0025] In this embodiment of the invention, the original interactive data stream is first obtained through the public interface of the social platform or through data cooperation. Then, the user's basic attribute information and historical behavior logs are separated from the data stream. Subsequently, the multi-source heterogeneous data is associated, integrated, and format standardized. Finally, structured social network interactive data containing user attribute features and historical information dissemination records is generated.

[0026] Step 102: Perform statistical analysis on the historical information dissemination records and the user attribute characteristics to generate personalized information dissemination parameters.

[0027] In this step, statistical analysis processing refers to using machine learning feature engineering methods to perform pattern mining and quantification on behavioral data; personalized information dissemination parameters refer to a set of probability parameters that reflect users' dissemination tendencies, obtained through calculation.

[0028] In this embodiment of the invention, firstly, the dissemination pattern features of historical information dissemination records are extracted; secondly, the social influence of user attribute features is quantified; then, the extracted dissemination pattern features and the quantified social influence index are weighted and fused together; finally, personalized information dissemination parameters reflecting the individual dissemination characteristics of users are generated.

[0029] Step 103: Configure the personalized information dissemination parameters into virtual user behavior entities corresponding to users of a preset real social network to generate a set of configured virtual user behavior entities.

[0030] In this step, the configuration operation refers to the parameter mapping process of assigning parameters to virtual entities; the preset real social network refers to the real network environment that serves as the simulation benchmark; the virtual user behavior entity refers to the computational agent that simulates real user behavior; and the set of configured virtual user behavior entities refers to the group of virtual entities whose parameters have been configured.

[0031] In this embodiment of the invention, firstly, virtual user behavior entities corresponding one-to-one with real social network users are created; secondly, a mapping table between virtual entities and real users is established; then, the information forwarding probability and content preference weight in the personalized information dissemination parameters are configured into the corresponding virtual entities; finally, the entity set state is updated to generate a set of configured virtual user behavior entities.

[0032] Step 104: Drive the configured set of virtual user behavior entities to simulate information propagation in a parallel simulation environment constructed based on the social network interaction data, and generate information flow paths.

[0033] In this step, the parallel simulation environment refers to a virtual simulation system that runs synchronously with the real network; the simulated information propagation operation refers to the process of information interaction between virtual entities driven by algorithms; and the information flow path refers to the trajectory sequence of information propagation in the simulated network.

[0034] In this embodiment of the invention, a simulated network environment is first constructed based on the user relationship topology in the social network interaction data. Then, the configured set of virtual user behavior entities is loaded into the simulation environment nodes. Subsequently, initial propagation information is injected and the virtual entities are driven to perform multiple rounds of interaction simulation. Finally, the information transmission trajectory is recorded to generate a complete information flow path.

[0035] Step 105: Analyze the information flow bottlenecks in the information flow path, connect the key transmission nodes connected to different virtual user behavior entities in the analysis results, and generate a risk transmission path.

[0036] In this step, the information flow bottleneck refers to a node or segment in the path where the propagation efficiency is significantly reduced; the analysis operation refers to the process of identifying key nodes using complex network analysis methods; the analysis result refers to a diagnostic report that includes a ranking of node importance; the key transmission node refers to a network node that has a key impact on information propagation; the connection operation refers to the graph calculation process of constructing risk transmission relationships between nodes; and the risk transmission path refers to the complete link of potential risk propagation.

[0037] In this embodiment of the invention, firstly, transmission delay points and congestion segments in the information flow path are detected; secondly, the node connection density and interaction frequency at the bottleneck location are analyzed; then, key transmission nodes connecting different virtual user behavior entities are identified; and finally, risk transmission paths are generated by reconstructing node connection relationships.

[0038] Step 106: Based on the risk transmission path, evaluate the impact of various information intervention strategies in the parallel simulation environment on the configured set of virtual user behavior entities, select the information intervention strategy whose impact meets the preset optimal impact condition as the final intervention plan, and deploy the final intervention plan in the preset real social network.

[0039] In this step, the evaluation operation refers to the process of quantifying the effect of the strategy through simulation experiments; the information intervention strategy refers to intervention methods including node suppression and competitive information delivery; the degree of impact refers to the indicator of the degree to which the strategy changes the risk transmission path; the preset optimal degree of impact condition refers to the pre-set evaluation criteria for the effect of the strategy; the final intervention plan refers to the best combination of strategies verified by simulation; and the deployment operation refers to the process of mapping the virtual environment strategy to the real network.

[0040] In this embodiment of the invention, firstly, multiple information intervention strategies are applied to the risk transmission path in a parallel simulation environment; secondly, the behavioral response data of the set of virtual user behavior entities are monitored; then, the inhibition effect coefficient of each strategy on risk transmission is calculated; and finally, the optimal strategy is selected and mapped to the corresponding node of the real social network.

[0041] For example, firstly, user profiles and post forwarding records are obtained through the Weibo platform developer interface. Secondly, feature extraction algorithms are used to analyze users' historical behavior to generate propagation tendency parameters. Then, virtual entities corresponding to real users are created and parameters are configured to form a simulated group. Next, hot topic information is injected into the constructed simulated network and the propagation path is observed. Then, key nodes and bottlenecks in the path are identified to form a risk transmission map. Finally, multiple intervention schemes are tested and the optimal strategy is selected for actual deployment on the Weibo platform.

[0042] This invention constructs a high-fidelity parallel simulation environment, transforming real social network data into virtual behavioral entity parameters. In the simulation environment, risk transmission paths are accurately deduced and various intervention strategies are tested. Finally, the validated optimal strategy is deployed to the real network, realizing the pre-prediction, in-process deduction, and precise intervention of social network risks, and significantly improving the accuracy and security of risk management.

[0043] To address the problem of raw social data being disorganized and difficult to use directly, this step receives data streams through a preset access point and processes them through structured transformation and time-series reassembly to generate standardized and usable social network interaction data. This invention provides a specific embodiment: Step 101, acquiring social network interaction data, wherein the social network interaction data includes user attribute characteristics and historical information dissemination records, specifically including the following steps: Step 111: Receive the raw interactive data stream from the preset real social network data source using the preset data access point.

[0044] In this step, the data source refers to the social platform server-side database system that provides raw data, including user information database and behavior log database; the raw interaction data stream refers to the set of unprocessed data containing basic user attributes, social relationships and behavioral events that are obtained in real time from the data source.

[0045] In this embodiment of the invention, a connection channel with the social platform server is first established through a pre-deployed application programming interface. Then, real-time data pushes from the platform server are received according to predetermined data specifications. Subsequently, garbled characters and missing values ​​generated during transmission are automatically corrected and completed. Finally, a complete and usable original interactive data stream is generated.

[0046] Step 112: Perform structured transformation on the user basic attribute data in the original interactive data stream to generate user attribute features.

[0047] In this step, user basic attribute data refers to the raw data items that reflect the user's basic characteristics, including fields such as age, gender, geographic distribution, and interest tags; structured transformation processing refers to the data processing process of converting non-standardized raw data into a unified format, including data type standardization, field alignment, and format normalization operations.

[0048] In this embodiment of the invention, firstly, user basic attribute data containing age, occupation, and interest tags in the original interactive data stream is identified; secondly, non-standardized attribute information is converted into a unified format according to a preset data structure template; then, data type verification is performed to ensure that the converted data conforms to the storage specifications of numerical and categorical data; finally, structured user attribute features are generated.

[0049] Step 113: Based on the time series of the original interactive data stream, reassemble the discrete propagation events in the original interactive data stream to generate historical information propagation records.

[0050] In this step, time series refers to a sequence of event data arranged in chronological order of occurrence, used to analyze the temporal regularity of behavioral patterns; discrete propagation events refer to scattered interactive behavior records such as likes, reposts, and comments generated by users on social networks; and reorganization processing refers to a data integration method that recombines scattered events according to time sequence and correlation, including event sorting, correlation establishment, and deduplication. In this embodiment of the invention, the timestamp information and event type identifier are first extracted from the original interactive data stream. Then, the scattered like, forward and comment events are sorted and combined according to the chronological order. Subsequently, the causal relationship between the events is established and duplicate records are removed. Finally, a continuous and complete historical information dissemination record is generated.

[0051] Step 114: Associate and integrate the user attribute features and the historical information dissemination records to generate social network interaction data.

[0052] In this step, the association integration operation refers to the data processing process of fusing data from different dimensions through identifier matching, including steps such as identifier mapping, data alignment, and redundancy elimination.

[0053] In this embodiment of the invention, a mapping table between user identifiers and event subject identifiers is first established. Then, user attribute features and historical information dissemination records are matched and aligned according to the identifiers. Subsequently, the matched data is subjected to integrity verification and redundancy elimination. Finally, multi-dimensional social network interaction data is generated.

[0054] This invention establishes a standardized data access and processing flow, transforming multi-source heterogeneous social network raw data into a structured interactive dataset. This ensures the real-time nature of data acquisition and the standardization of the processing, providing high-quality data input for subsequent parallel simulation environments and effectively improving the accuracy and reliability of risk transmission simulation.

[0055] To improve the accuracy of quantifying user dissemination characteristics, this step extracts historical dissemination pattern features and calculates a social influence index, then integrates these features to generate personalized parameters reflecting an individual's dissemination tendencies. This invention provides a specific embodiment where step 102 involves statistically analyzing the historical information dissemination records and user attribute features to generate personalized information dissemination parameters, specifically including the following steps: Step 201: Extract the propagation pattern features from the historical information propagation records, perform statistics on the propagation pattern features, and generate the user's historical propagation pattern.

[0056] In this step, the propagation pattern characteristics refer to the characteristic indicators reflecting the characteristics of user propagation behavior extracted from historical propagation data, including dimensions such as propagation speed, propagation breadth, and propagation duration; statistical operations refer to the processing of quantitative analysis of characteristic data, including methods such as frequency calculation, distribution analysis, and correlation measurement; user historical propagation patterns refer to the models reflecting the patterns of user historical propagation behavior obtained through statistical analysis, which are used to predict future propagation behavior.

[0057] In this embodiment of the invention, the characteristics of information dissemination speed, dissemination range and dissemination duration are first identified from historical information dissemination records. Then, frequency distribution statistics and regularity analysis are performed on these dissemination pattern characteristics. Subsequently, the proportion and interrelationship of different types of dissemination patterns are calculated. Finally, a user historical dissemination pattern reflecting the user's historical dissemination behavior is generated.

[0058] Step 202: Perform weighted calculation on the number of followers and interaction frequency data in the user attribute features to generate the user's social influence index.

[0059] In this step, the number of followers refers to the total number of people who follow the user on the social network; the interaction frequency data refers to the frequency of interaction between the user and other users, including the occurrence rate of interaction behaviors such as likes, comments and reposts; the weighted calculation processing refers to the mathematical processing method of assigning different importance weights to different indicators and performing comprehensive calculations; the user social influence index is a quantitative indicator obtained through weighted calculation, used to measure the scope and intensity of the user's influence on the social network.

[0060] In this embodiment of the invention, firstly, the number of followers and interaction frequency data in the user attribute features are extracted; secondly, a higher weight value is assigned to the number of followers and a moderate weight value is assigned to the interaction frequency; then, the weighted number of followers and the weighted interaction frequency are added together; finally, a quantified user social influence index is generated.

[0061] Step 203: Combine the user's historical propagation pattern and the user's social influence index to generate a propagation tendency parameter.

[0062] In this step, the fusion calculation operation refers to the calculation process of integrating multiple features or indicators, including feature vectorization and weighted fusion; the propagation tendency parameter refers to the set of parameters obtained by combining the user's historical propagation patterns and social influence, reflecting the user's comprehensive propagation characteristics.

[0063] In this embodiment of the invention, the user's historical propagation pattern is first converted into a numerical feature vector, then the user's social influence index is standardized, then the two feature vectors are weighted and fused, and finally a comprehensive propagation tendency parameter is generated.

[0064] Step 204: Extract the information forwarding probability and content preference weight from the propagation tendency parameters.

[0065] In this step, the information forwarding probability refers to the numerical value of the likelihood that a user will forward a specific piece of information; the content preference weight refers to the quantitative indicator of a user's preference for a certain type of content.

[0066] In this embodiment of the invention, firstly, parameters related to information propagation behavior are separated from the propagation tendency parameters; secondly, values ​​reflecting the possibility of forwarding are identified from these parameters as information forwarding probabilities; subsequently, values ​​reflecting the intensity of content preference are identified as content preference weights; and finally, the extraction of the two key parameters is completed.

[0067] Step 205: Configure the information forwarding probability and the content preference weight to each virtual user behavior entity corresponding to the user of the preset real social network, and generate personalized information dissemination parameters.

[0068] In this embodiment of the invention, firstly, a correspondence table between virtual user behavior entities and real users is established; secondly, the information forwarding probability is assigned to the propagation behavior parameters of the corresponding virtual entity; subsequently, the content preference weight is assigned to the content selection parameters of the corresponding virtual entity; and finally, personalized information propagation parameters containing complete parameter settings are generated.

[0069] This invention generates a parameter system that accurately reflects the characteristics of user dissemination by deeply analyzing and integrating users' historical dissemination behavior and social influence. This provides accurate behavioral configuration basis for virtual user behavior entities and significantly improves the realism of information dissemination simulation and the accuracy of risk transmission inference in parallel simulation environments.

[0070] To address the issue of low matching rates between virtual entities and real user behavior, this step establishes a precise correspondence and assigns personalized parameters to virtual entities, generating a high-fidelity set of virtual user behavior entities. This invention provides a specific embodiment where step 103 involves configuring the personalized information propagation parameters into virtual user behavior entities corresponding to users on a preset real social network, thereby generating a configured set of virtual user behavior entities. This specifically includes the following steps: Step 301: Create virtual user behavior entities that correspond one-to-one with users of each preset real social network.

[0071] In this step, the creation operation refers to the process of generating a corresponding virtual entity based on real user information, including steps such as entity initialization, attribute setting, and identifier allocation.

[0072] In this embodiment of the invention, the number of virtual entities to be created is first determined based on the number of users in the real social network. Then, a unique identifier is assigned to each real user and a corresponding virtual behavior entity is created. Subsequently, a basic attribute framework is initialized for each virtual entity. Finally, the creation process of all virtual user behavior entities is completed.

[0073] Step 302: Combine the virtual user behavior entities to generate an initial set of virtual user behavior entities.

[0074] In this step, the combination operation refers to the process of organizing and integrating multiple virtual entities according to a specific structure, including operations such as grouping, connecting, and establishing storage structures; the initial set of virtual user behavior entities refers to a group of virtual entities that only contains the basic attribute framework and has not yet been configured with propagation parameters.

[0075] In this embodiment of the invention, the created virtual user behavior entities are first grouped according to the group structure of the real social network. Then, the connection relationship between the entities is established to reflect the social network topology of the real users. Subsequently, all the grouped entities are integrated into a unified storage structure, and finally, an initial set of virtual user behavior entities is generated.

[0076] Step 303: Establish the correspondence between each of the virtual user behavior entities and each of the preset real social network users.

[0077] In this step, the correspondence refers to the two-way mapping relationship established between virtual entities and real users, which is used to ensure the accuracy of parameter configuration.

[0078] In this embodiment of the invention, the unique identity information of real social network users is first obtained, then a corresponding real user identifier is assigned to each virtual user behavior entity, then a bidirectional mapping table is established to record the correspondence between each virtual entity and the real user, and finally the establishment of all correspondences is completed.

[0079] Step 304: Based on the correspondence, allocate the information forwarding probability in the personalized information propagation parameters to each of the virtual user behavior entities to generate a probability configuration result.

[0080] In this step, the allocation operation refers to the process of setting parameter values ​​into virtual entities according to the corresponding relationship, including steps such as parameter reading, matching allocation, and value setting; the probability configuration result refers to the data set that records the forwarding probability allocation of all virtual entity information.

[0081] In this embodiment of the invention, the information forwarding probability data in the personalized information propagation parameters is first read, then the probability value that should be allocated to each virtual entity is determined according to the established correspondence, then the probability value is configured into the behavior parameters of the corresponding virtual entity, and finally a probability configuration result containing the probability configuration information of all virtual entities is generated.

[0082] Step 305: Based on the correspondence, allocate the content preference weights in the personalized information dissemination parameters to each of the virtual user behavior entities to generate weight configuration results.

[0083] In this step, the weight configuration result refers to the data set that records the content preference weight settings of all virtual entities; In this embodiment of the invention, firstly, the content preference weight data in the personalized information dissemination parameters is obtained; secondly, the weight values ​​are assigned to the corresponding virtual user behavior entities according to the correspondence; then, the weight values ​​are set in the content selection module of the virtual entity; and finally, a weight configuration result recording the weight configuration of all entities is generated.

[0084] Step 306: Update the initial set of virtual user behavior entities based on the probability configuration result and the weight configuration result to generate a configured set of virtual user behavior entities.

[0085] In this step, the update operation refers to the process of modifying the parameters of an existing entity set using new configuration information, including operations such as parameter reading, value updating, and state refreshing.

[0086] In this embodiment of the invention, the probability allocation information in the probability configuration result is read first, the weight setting information in the weight configuration result is obtained second, and then the parameter settings of each entity in the initial virtual user behavior entity set are updated using these configuration information. Finally, a configured virtual user behavior entity set containing complete parameter configuration is generated.

[0087] By establishing a precise correspondence between virtual entities and real users, this invention enables the accurate configuration of personalized information dissemination parameters. This ensures that the virtual user behavior entities in the parallel simulation environment can accurately reflect the dissemination characteristics of actual users, laying a solid foundation for the accurate deduction of subsequent risk transmission paths.

[0088] To accurately simulate the information propagation process in social networks, this step constructs a parallel simulation environment and drives virtual entities to perform interactive simulations, extracting and integrating the data to generate a complete information flow path. This invention provides a specific embodiment: Step 104, driving the configured set of virtual user behavior entities in the parallel simulation environment constructed based on the social network interaction data to simulate information propagation and generate an information flow path, specifically includes the following steps: Step 401: Construct a parallel simulation environment using the user relationship topology in the social network interaction data.

[0089] In this step, user relationship topology refers to the network structure representation of social connections between users, including connection patterns such as follow relationships, friend relationships, and interaction relationships.

[0090] In this embodiment of the invention, firstly, the follow relationships and friend connection data between users are extracted from social network interaction data; secondly, a network topology is constructed based on these connection relationships; then, the basic attributes and connection weights of network nodes are set; and finally, a parallel simulation environment capable of simulating the structure of a real social network is generated.

[0091] Step 402: Load the configured set of virtual user behavior entities into the nodes of the parallel simulation environment and generate node loading results.

[0092] In this step, the loading operation refers to the process of placing virtual entities into the simulation environment, including steps such as location matching, entity placement, and state recording; a node refers to a basic unit in the network topology, representing a connection point or location point; and the node loading result refers to a data set that records the location information of virtual entities in the simulation environment.

[0093] In this embodiment of the invention, firstly, a set of configured virtual user behavior entities is obtained; secondly, the location identifiers of each node in the parallel simulation environment are identified; then, each virtual user behavior entity is placed into the corresponding network node; and finally, a node loading result recording the location information of all entities is generated.

[0094] Step 403: Input the preset initial propagation information into the parallel simulation environment to trigger the virtual user behavior entities configured in the node loading result to perform simulated interaction behavior and generate simulated interaction behavior data.

[0095] In this step, the preset initial propagation information refers to the initial information content that is set in advance to trigger the propagation simulation; the simulated interactive behavior refers to the information propagation and interactive activities carried out by the virtual entity in the simulation environment; and the simulated interactive behavior data refers to the data set that records all interactive behaviors and propagation processes.

[0096] In this embodiment of the invention, firstly, a preset initial propagation information content is prepared; secondly, the information is input into a specific starting node of the parallel simulation environment; then, the virtual user behavior entity in the node loading result is triggered to perform information interaction and propagation behavior according to its configuration parameters; finally, simulated interaction behavior data containing all interaction records is generated.

[0097] Step 404: Determine the transmission direction of the preset initial propagation information based on each configured virtual user behavior entity to obtain information transmission direction data; In this step, the transmission direction refers to the path and direction of information propagation in the network; the information transmission direction data refers to the data set that records the information propagation direction.

[0098] In this embodiment of the invention, the information forwarding probability parameters of each configured virtual user behavior entity are first analyzed, then the potential direction of information propagation is calculated based on these probability values, then the transmission path selection of information at each network node is determined, and finally detailed information transmission direction data is generated.

[0099] Step 405: Based on the information transmission direction data, extract multiple initial information propagation paths from the information transmission trajectory of the simulated interaction behavior data.

[0100] In this step, the information transmission trajectory refers to the path record of information propagation in the network; the initial information propagation path refers to the discrete propagation route record initially obtained by simulating the information propagation process in a parallel simulation environment.

[0101] In this embodiment of the invention, the information movement records in the simulated interactive behavior data are first analyzed, the information propagation trajectory is identified based on the information transmission direction data, multiple independent initial information propagation paths are then extracted from these trajectories, and finally the preliminary extraction of all paths is completed.

[0102] Step 406: Integrate the topology of each initial information propagation path to generate an information flow path.

[0103] In this step, topology refers to the structural characteristics of the connection relationships between nodes in the network; integration operation refers to the process of merging and optimizing multiple paths or structures.

[0104] In this embodiment of the invention, the network topology characteristics of each initial information propagation path are first analyzed, then the intersections and overlapping segments between paths are identified, then these paths are merged and optimized according to the topological relationship, and finally a complete information flow path is generated.

[0105] This invention constructs a high-fidelity parallel simulation environment and drives configured virtual entities within it to simulate information propagation. This accurately recreates the flow of information in social networks, generates detailed information flow paths, provides a reliable data foundation for subsequent risk transmission analysis, and significantly improves the accuracy of risk prediction.

[0106] To accurately identify key nodes and paths of risk transmission, this step constructs and verifies a reliable risk transmission path by detecting information flow bottlenecks and analyzing node connections. This invention provides a specific embodiment: Step 105 involves analyzing information flow bottlenecks in the information flow path, connecting key transmission nodes linked to different virtual user behavior entities in the analysis results, and generating a risk transmission path. This specifically includes the following steps: Step 501: Detect information transmission delay points in the information flow path to identify the bottleneck location of information flow from the detection results.

[0107] In this step, the information transmission delay point refers to the specific location where the transmission time of information is abnormally extended during the propagation process, which is usually identified by calculating the timestamp difference; the detection operation refers to the process of identifying and locating abnormal points in the information flow path, including time monitoring, difference calculation and abnormal identification steps; the detection result refers to the data output that records the information of all identified abnormal points; the information flow bottleneck location refers to the key area where the information transmission efficiency is significantly reduced, which is usually formed by the clustering of multiple delay points.

[0108] In this embodiment of the invention, the information transmission timestamps of each node in the information flow path are first monitored, the transmission time difference between adjacent nodes is calculated, abnormal points where the transmission time difference exceeds the normal range are identified, and finally the bottleneck position of information flow is determined from these abnormal points.

[0109] Step 502: Analyze the node connection relationships at the bottleneck location of the information flow to determine the hub node connecting different virtual user behavior entities.

[0110] In this step, node connection relationship refers to data describing the connection status between network nodes, including information such as connection direction, connection strength and connection type; analysis operation refers to the process of parsing and evaluating node connection data, aiming to discover important connection patterns; hub node refers to a key node that connects multiple different network areas and plays a bridging role in the dissemination of network information.

[0111] In this embodiment of the invention, firstly, the connection relationship data of all nodes at the bottleneck position of information flow is obtained; secondly, the connection density and connection strength of these nodes with other virtual user behavior entities are analyzed; then, the cross nodes that simultaneously connect multiple different virtual user behavior entities are identified; and finally, these cross nodes are determined as hub nodes.

[0112] Step 503: Mark each key transmission node in the hub node to generate a key node set.

[0113] In this step, the tagging operation refers to the process of adding identification tags to important nodes to facilitate subsequent tracking and analysis; the key node set refers to the data set containing all important nodes, which is used for subsequent path construction.

[0114] In this embodiment of the invention, the importance of each hub node in information dissemination is first assessed, then key transmission nodes are selected according to a preset importance threshold, then special identifiers are added to these key transmission nodes, and finally a set of key nodes containing all key transmission nodes is generated.

[0115] Step 504: Establish the connection relationship between each key transmission node according to the set of key nodes, traverse each connection relationship, and generate multiple initial risk transmission paths.

[0116] In this step, the connection relationship refers to the description of the association between nodes, including direct and indirect connections; the initial risk transmission path refers to the preliminary risk propagation route, which has not yet been verified for completeness.

[0117] In this embodiment of the invention, the connection information of each key transmission node in the key node set is first read, then the direct and indirect connection relationships between these nodes are established, then all possible connection combinations are traversed, and finally multiple initial risk transmission paths are generated.

[0118] Step 505: Perform path integrity verification on each of the initial risk transmission paths, and take the initial risk transmission path corresponding to the verification result that meets the preset pass conditions as the risk transmission path.

[0119] In this step, the route integrity verification operation refers to the process of checking and evaluating the feasibility of the route, including continuity checks and feasibility verification; the preset pass conditions refer to the standard requirements for the route verification to pass; and the verification result refers to the qualification assessment conclusion obtained after the route verification.

[0120] In this embodiment of the invention, the continuity of node connections in each initial risk transmission path is first checked, the feasibility of information propagation in the path is verified, the qualification of each path is evaluated according to a preset integrity standard, and finally the path that meets the standard is determined as the final risk transmission path.

[0121] This invention establishes a path model that accurately reflects the risk transmission pattern by accurately identifying and analyzing bottlenecks and key nodes in the information flow path. The feasibility and accuracy of the path are ensured through integrity verification, providing a reliable basis for subsequent risk assessment and intervention strategy formulation, and significantly improving the accuracy and practicality of risk transmission prediction.

[0122] To improve the effectiveness and safety of intervention strategies, this step evaluates the impact of multiple strategies in a simulation environment, selects the optimal solution, and maps it to a real social network. This invention provides a specific embodiment: Step 106, based on the risk transmission path, evaluates the impact of multiple information intervention strategies in the parallel simulation environment on the configured set of virtual user behavior entities, selects the information intervention strategy whose impact meets the preset optimal impact condition as the final intervention solution, and deploys the final intervention solution to the preset real social network, specifically including the following steps: Step 601: Execute various information intervention strategies in the parallel simulation environment at each of the key transmission nodes on the risk transmission path, and generate strategy application results.

[0123] In this step, the strategy application results refer to the effect data recorded after implementing various information intervention strategies in a parallel simulation environment, including information such as node response status and network propagation changes.

[0124] In this embodiment of the invention, firstly, all key transmission nodes on the risk transmission path are identified; secondly, multiple preset information intervention strategies are sequentially implemented on these nodes in a parallel simulation environment; then, the response status of the nodes and changes in network propagation under the action of each strategy are recorded; finally, a strategy application result containing all strategy execution statuses is generated.

[0125] Step 602: Monitor the behavioral change characteristics of the set of virtual user behavior entities configured in the policy application results, so as to collect behavioral change data from the monitoring results.

[0126] In this step, behavioral change characteristics refer to the behavioral pattern changes exhibited by virtual user entities under the influence of policies, including changes in forwarding frequency and content preferences; monitoring operations refer to the process of continuously observing and recording the behavior of virtual entities, including data collection and feature extraction steps; monitoring results refer to the raw observation data records obtained from the monitoring operations; and behavioral change data refer to the structured behavioral change records that have been organized for subsequent analysis and processing. In this embodiment of the invention, firstly, for the set of virtual user behavior entities configured in the policy application results, the changes in their information forwarding frequency and content selection preferences are continuously monitored; secondly, data on the changes in the interaction relationships and propagation range between entities are collected; then, these changes are organized into structured records; and finally, detailed behavioral change data is generated.

[0127] Step 603: Analyze the behavioral change data and calculate the impact index of each information intervention strategy based on the analysis results.

[0128] In this step, the analysis operation refers to the process of parsing and evaluating behavioral change data, aiming to discover patterns of change and trends in effects; the analysis result refers to the conclusive data output obtained after the analysis operation; and the impact degree index refers to a numerical indicator that quantifies the magnitude of the effect of the information intervention strategy, reflecting the strength of the strategy's impact on risk transmission.

[0129] In this embodiment of the invention, firstly, trend analysis and pattern recognition are performed on behavioral change data; secondly, the degree of change in risk transmission path under the action of each information intervention strategy is calculated; then, the effectiveness of the strategy in blocking or guiding information flow is quantified; and finally, comparable impact indicators are generated.

[0130] Step 604: Compare the various influence degree indicators to select the influence degree indicator that meets the preset optimal influence degree conditions as the target influence degree indicator.

[0131] In this step, the comparison operation refers to the process of comparing and analyzing multiple indicators, including sorting and screening steps; the target impact index refers to the best effect index determined after comparison and screening. In this embodiment of the invention, a comparative analysis framework for each influence degree indicator is first established. Then, the indicators are sorted and screened according to the preset optimal influence degree conditions. Subsequently, the best effect indicator that meets the conditions is identified and finally determined as the target influence degree indicator.

[0132] Step 605: Select the information intervention strategy corresponding to the target impact index as the final intervention plan.

[0133] In this embodiment of the invention, firstly, a correspondence table between the degree of influence index and the information intervention strategy is established; secondly, the corresponding strategy identifier is found according to the determined target degree of influence index; then, the strategy is selected as the final intervention plan; and finally, the plan selection process is completed.

[0134] Step 606: Map the final intervention plan to the preset real social network to complete the deduction and intervention of the risk transmission path.

[0135] In this step, the mapping operation refers to the process of converting intervention schemes in the simulation environment into executable instructions for the real network.

[0136] In this embodiment of the invention, the parameter settings in the final intervention plan are first converted into execution instructions of the real social network. Then, these instructions are mapped to the corresponding user nodes in the real network. Subsequently, intervention measures are implemented and the execution status is confirmed. Finally, the entire risk transmission path is deduced and the intervention process is completed.

[0137] This invention, through testing various information intervention strategies in a parallel simulation environment and quantifying their impact, enables the scientific and accurate selection of the optimal intervention plan. The validated plan is then securely deployed to real social networks, achieving pre-emptive projection and precise intervention of risk transmission, thus greatly improving the effectiveness and security of risk management.

[0138] Figure 2 This invention provides a schematic diagram of a risk transmission path deduction and intervention system based on a parallel simulation environment, as shown in the embodiment of the invention. Figure 2 As shown, the system includes: The acquisition module 21 is used to acquire social network interaction data, wherein the social network interaction data includes user attribute characteristics and historical information dissemination records; Processing module 22 is used to perform statistical analysis and processing on the historical information dissemination records and the user attribute characteristics to generate personalized information dissemination parameters; Configuration module 23 is used to configure the personalized information dissemination parameters into virtual user behavior entities corresponding to users of a preset real social network, so as to generate a set of configured virtual user behavior entities; The driving module 24 is used to drive the configured set of virtual user behavior entities to simulate information propagation and generate information flow paths in a parallel simulation environment constructed based on the social network interaction data. The connection module 25 is used to analyze the information flow bottlenecks in the information flow path, connect the key transmission nodes connected to different virtual user behavior entities in the analysis results, and generate a risk transmission path. Evaluation module 26 is used to evaluate the impact of various information intervention strategies in the parallel simulation environment on the configured set of virtual user behavior entities based on the risk transmission path, select the information intervention strategy whose impact meets the preset optimal impact condition as the final intervention plan, and deploy the final intervention plan into the preset real social network.

[0139] Figure 2 The aforementioned risk transmission path simulation and intervention system based on a parallel simulation environment can execute... Figure 1 The implementation principle and technical effects of the risk transmission path deduction and intervention method based on a parallel simulation environment described in the illustrated embodiment will not be repeated here. The specific methods by which each module and unit of the risk transmission path deduction and intervention system based on a parallel simulation environment in the above embodiments have been described in detail in the embodiments related to this method, and will not be elaborated upon here.

[0140] In one possible design, Figure 2 The risk transmission path simulation and intervention system shown in the embodiment can be implemented as a computing device, such as... Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32; The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are invoked and executed by the processing component 32.

[0141] The processing component 32 is used to: acquire social network interaction data, wherein the social network interaction data includes user attribute characteristics and historical information dissemination records; perform statistical analysis on the historical information dissemination records and the user attribute characteristics to generate personalized information dissemination parameters; configure the personalized information dissemination parameters in virtual user behavior entities corresponding to users of a preset real social network to generate a set of configured virtual user behavior entities; drive the set of configured virtual user behavior entities to simulate information dissemination in a parallel simulation environment constructed based on the social network interaction data to generate information flow paths; analyze the information flow bottlenecks in the information flow paths, connect the key transmission nodes connected to different virtual user behavior entities in the analysis results to generate risk transmission paths; based on the risk transmission paths, evaluate the degree of influence of various information intervention strategies in the parallel simulation environment on the set of configured virtual user behavior entities, select the information intervention strategy whose degree of influence meets the preset optimal degree of influence conditions as the final intervention scheme, and deploy the final intervention scheme to the preset real social network.

[0142] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above-described method. Alternatively, the processing component may be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above-described method.

[0143] Storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0144] Of course, computing devices may also include other components, such as input / output interfaces, display components, communication components, etc.

[0145] Input / output interfaces provide interfaces between processing components and peripheral interface modules, which can be output devices, input devices, etc.

[0146] The communication components are configured to facilitate wired or wireless communication between computing devices and other devices.

[0147] The computing device can be a physical device or an elastic computing host provided by a cloud computing platform. In this case, the computing device can refer to a cloud server, and the aforementioned processing components, storage components, etc., can be basic server resources rented or purchased from the cloud computing platform.

[0148] This invention also provides a computer storage medium storing a computer program, which, when executed by a computer, can perform the above-described functions. Figure 1 The embodiment shown is a method for risk transmission path deduction and intervention based on a parallel simulation environment.

[0149] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0150] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0151] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0152] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for risk transmission path deduction and intervention based on a parallel simulation environment, characterized in that, include: Acquire social network interaction data, wherein the social network interaction data includes user attribute characteristics and historical information dissemination records; wherein, user attribute characteristics refer to static attribute data including user age, gender, region, and interest tags; and historical information dissemination records refer to time series data reflecting the dissemination behavior of users in the past, such as posting, forwarding, and commenting. Statistical analysis and processing are performed on the historical information dissemination records and the user attribute characteristics to generate personalized information dissemination parameters; The personalized information dissemination parameters are configured in virtual user behavior entities corresponding to users of a preset real social network to generate a set of configured virtual user behavior entities. In a parallel simulation environment constructed based on the social network interaction data, the configured set of virtual user behavior entities is driven to simulate information propagation and generate information flow paths. The information flow bottlenecks in the information flow path are analyzed, and the key transmission nodes connected to different virtual user behavior entities in the analysis results are connected to generate risk transmission paths. Based on the risk transmission path, the impact of various information intervention strategies in the parallel simulation environment on the configured set of virtual user behavior entities is evaluated. The information intervention strategy whose impact meets the preset optimal impact condition is selected as the final intervention plan, and the final intervention plan is deployed in the preset real social network. The information flow bottlenecks in the information flow path are analyzed, and the key transmission nodes connected to different virtual user behavior entities in the analysis results are connected to generate a risk transmission path, including: Information transmission delay points in the information flow path are detected to identify the bottleneck locations of information flow from the detection results; Analyze the node connection relationships at the bottleneck locations of the information flow to determine the hub nodes that connect different virtual user behavior entities; Each key transmission node in the hub node is marked to generate a key node set; Based on the set of key nodes, establish the connection relationship between each key transmission node, traverse each connection relationship, and generate multiple initial risk transmission paths; Each initial risk transmission path is verified for path integrity, and the initial risk transmission path corresponding to the verification result that meets the preset pass conditions is taken as the risk transmission path.

2. The method according to claim 1, characterized in that, Acquire social network interaction data, wherein the social network interaction data includes user attribute characteristics and historical information dissemination records, including: Utilize preset data access points to receive raw interactive data streams from preset real social network data sources; The user basic attribute data in the original interactive data stream is subjected to structured transformation processing to generate user attribute features; Based on the time series of the original interactive data stream, the discrete propagation events in the original interactive data stream are recombined to generate historical information propagation records; The user attribute features and the historical information dissemination records are linked and integrated to generate social network interaction data.

3. The method according to claim 1, characterized in that, Statistical analysis is performed on the historical information dissemination records and the user attribute characteristics to generate personalized information dissemination parameters, including: The propagation pattern features are extracted from the historical information propagation records, and the propagation pattern features are statistically analyzed to generate the user's historical propagation pattern. The user's social influence index is generated by performing a weighted calculation on the number of followers and interaction frequency data in the user attribute features. The user's historical propagation pattern and the user's social influence index are fused and calculated to generate a propagation tendency parameter; Extract the information forwarding probability and content preference weight from the propagation tendency parameters; The information forwarding probability and the content preference weight are configured into each virtual user behavior entity corresponding to users of a preset real social network to generate personalized information dissemination parameters.

4. The method according to claim 1, characterized in that, The personalized information dissemination parameters are configured into virtual user behavior entities corresponding to users of a preset real social network to generate a configured set of virtual user behavior entities, including: Create virtual user behavior entities that correspond one-to-one with users in each preset real social network; The virtual user behavior entities are combined to generate an initial set of virtual user behavior entities; Establish a correspondence between each of the virtual user behavior entities and each of the preset real social networks; Based on the correspondence, the information forwarding probability in the personalized information dissemination parameters is allocated to each of the virtual user behavior entities to generate a probability configuration result; Based on the correspondence, the content preference weights in the personalized information dissemination parameters are allocated to each of the virtual user behavior entities to generate weight configuration results; Based on the probability configuration result and the weight configuration result, the initial virtual user behavior entity set is updated to generate a configured virtual user behavior entity set.

5. The method according to claim 1, characterized in that, In a parallel simulation environment constructed based on the social network interaction data, the configured set of virtual user behavior entities is driven to simulate information propagation and generate information flow paths, including: A parallel simulation environment is constructed using the user relationship topology in the aforementioned social network interaction data; The configured set of virtual user behavior entities is loaded into the nodes of the parallel simulation environment to generate node loading results; The preset initial propagation information is input into the parallel simulation environment to trigger the virtual user behavior entities configured in the node loading result to perform simulated interaction behavior and generate simulated interaction behavior data. The transmission direction of the preset initial propagation information is determined based on each configured virtual user behavior entity to obtain information transmission direction data; Based on the information transmission direction data, multiple initial information propagation paths are extracted from the information transmission trajectory of the simulated interaction behavior data; The topological structures in each of the initial information propagation paths are integrated to generate an information flow path.

6. The method according to claim 1, characterized in that, Based on the risk transmission path, the impact of various information intervention strategies in the parallel simulation environment on the configured set of virtual user behavior entities is evaluated. The information intervention strategy whose impact meets the preset optimal impact condition is selected as the final intervention plan. The final intervention plan is then deployed to the preset real social network, including: At each of the key transmission nodes along the risk transmission path, various information intervention strategies in the parallel simulation environment are executed to generate strategy application results; The behavioral change characteristics of the set of virtual user behavior entities configured in the application results of the strategy are monitored in order to collect behavioral change data from the monitoring results; The behavioral change data are analyzed, and the impact index of each information intervention strategy is calculated based on the analysis results. The various influence degree indicators are compared to select the influence degree indicator that meets the preset optimal influence degree condition as the target influence degree indicator. Select the information intervention strategy corresponding to the target impact level index as the final intervention plan; The final intervention plan is mapped onto the preset real social network to complete the deduction and intervention of the risk transmission path.

7. A risk transmission path deduction and intervention system based on a parallel simulation environment, used in the risk transmission path deduction and intervention method based on a parallel simulation environment as described in any one of claims 1 to 6, characterized in that, include: The acquisition module is used to acquire social network interaction data, wherein the social network interaction data includes user attribute characteristics and historical information dissemination records; The processing module is used to perform statistical analysis and processing on the historical information dissemination records and the user attribute characteristics to generate personalized information dissemination parameters; The configuration module is used to configure the personalized information dissemination parameters into virtual user behavior entities corresponding to users of a preset real social network, so as to generate a set of configured virtual user behavior entities. The driving module is used to drive the configured set of virtual user behavior entities to simulate information propagation and generate information flow paths in a parallel simulation environment constructed based on the social network interaction data. The connection module is used to analyze the information flow bottlenecks in the information flow path, connect the key transmission nodes connected to different virtual user behavior entities in the analysis results, and generate a risk transmission path. The evaluation module is used to evaluate the impact of various information intervention strategies in the parallel simulation environment on the configured set of virtual user behavior entities based on the risk transmission path, select the information intervention strategy whose impact meets the preset optimal impact condition as the final intervention plan, and deploy the final intervention plan into the preset real social network.

8. A computing device, characterized in that, It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component to implement the risk transmission path deduction and intervention method based on a parallel simulation environment as described in any one of claims 1 to 6.

9. A computer storage medium, characterized in that, The system contains a computer program that, when executed by a computer, implements a risk transmission path deduction and intervention method based on a parallel simulation environment as described in any one of claims 1 to 6.

Images