Multi-agent-based crowd belief evolution and intervention evaluation method and system

Abstract

The invention discloses a multi-agent-based crowd belief evolution and intervention evaluation method and system, and the method comprises the steps: endowing each agent with a set of multi-dimensional attributes, and quantifying the belief of the agent for a predetermined issue into a multi-dimensional continuous vector; constructing a simulation information ecological system environment containing a plurality of information sources; defining interaction rules between the intelligent agents and between the intelligent agents and environment information, wherein the interaction rules are based on sociology psychology and a propaganda science model; packaging a plurality of intervention strategies to be evaluated into a pluggable standardized module; a non-intervention baseline scene and a plurality of intervention strategy scenes are operated in a simulation environment, belief evolution data of group and individual levels are tracked and recorded in real time, and different intervention strategies are quantitatively evaluated and sorted by comparing key indexes. According to the method, the formation and evolution of the crowd belief can be truly simulated in the virtual environment, and various intervention strategies are evaluated, so that reliable support is provided for scientific decision making.

Description

Technical Field

[0001] This invention relates to the interdisciplinary field of computational sociology, artificial intelligence, and information intervention technology, and in particular to a method and system for evaluating and intervening in population belief evolution based on multi-agent intelligence. Background Technology

[0002] In fields such as public opinion management, combating misinformation, and risk control, predicting the evolution of population beliefs and evaluating the effectiveness of different intervention strategies are of great significance. With the rapid development of social media and self-media, the speed and scope of information dissemination have significantly expanded, leading to a more complex and dynamic process in the formation and evolution of group beliefs, posing serious challenges to decision-making by relevant organizations, enterprises, and research institutions.

[0003] In existing technologies, common methods mainly include the following categories:

[0004] One type of method is based on macroscopic differential equations and system dynamics models. For example, it abstracts the entire population into several group states and uses variants of epidemic models such as SIR (Susceptible-Infected-Recovered) to describe the spread of rumors or the evolution of public opinion. These models are characterized by their computational simplicity and strong analyzability, but they have obvious limitations: they assume that the population consists of homogeneous individuals, making it difficult to reflect the psychological differences and heterogeneous behaviors among individuals, resulting in a large deviation between simulation results and reality.

[0005] Another approach relies on social surveys and statistical analysis, using methods such as questionnaires and interviews to collect data on public attitudes, and then combining this with statistical techniques such as regression models and correlation analysis to predict trends. While this method can provide some empirical support, its data collection and analysis process is time-consuming and labor-intensive, and the results often exhibit significant lag, failing to meet the needs for real-time prediction and rapid response to sudden public opinion events.

[0006] In recent years, some studies have attempted to introduce agent-based modeling (ABM) methods to model opinion formation and public opinion diffusion. However, most existing ABM models only use a single scalar to represent an individual's opinion, and the update rules are often simple threshold or averaging mechanisms, lacking a characterization of the complex psychological mechanisms in real society. Furthermore, these models generally fail to effectively integrate with multi-source information data and lack a systematic intervention strategy evaluation module, thus limiting the reliability and application value of the simulation results.

[0007] In summary, existing technologies generally suffer from the following shortcomings when dealing with the problem of population belief evolution: insufficient modeling of individual heterogeneity and psychological characteristics; lack of simulation frameworks that can integrate multi-source information environments; lack of modularly injectable and systematically evaluated information intervention strategies; and weak real-time performance and interpretability of prediction and evaluation results.

[0008] Therefore, how to establish a simulation method and system that can reflect the differences in individual psychological and social attributes and support the injection of multi-dimensional information intervention strategies and quantitative evaluation has become a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0009] To address the aforementioned technical problems in the existing technology, this invention proposes a multi-agent-based method and system for population belief evolution and intervention assessment.

[0010] According to a first aspect of the present invention, a method for evaluating population belief evolution and intervention based on multi-agent systems is proposed, comprising:

[0011] S1: Assign each agent a set of multidimensional attributes, including sociodemographic attributes, social network attributes, and psychological cognitive attributes, and quantify the agent's belief in a predetermined issue into a multidimensional continuous vector;

[0012] S2: Construct a simulated information ecosystem environment containing multiple information sources, and define the credibility, dissemination intensity, and stance bias attributes of each information source;

[0013] S3: Defines the interaction rules between agents and between agents and environmental information. The interaction rules are based on sociological psychology and communication models.

[0014] S4: Encapsulate multiple intervention strategies to be evaluated into pluggable standardized modules. Intervention strategies include content strategies, channel strategies, timing strategies, and target strategies.

[0015] S5: Run baseline scenarios without intervention and multiple intervention strategy scenarios in a simulation environment, track and record belief evolution data at the group and individual levels in real time, and quantitatively evaluate and rank different intervention strategies by comparing key indicators.

[0016] In some specific embodiments, belief quantization quantifies the agent's belief in a predetermined issue into a multidimensional continuous vector B = [b1, b2, ..., bn], where each dimension represents an aspect of the belief and its value varies continuously in the interval [-1, 1].

[0017] In some specific embodiments, the interaction rules include: the agent is influenced by its social network neighbors, and the influence weight is determined by the closeness of the relationship and the influence of the other party; after receiving information, the degree of belief update of the agent depends on the credibility of the information source, the consistency between the information content and its own existing beliefs, and its own cognitive reflection ability; the agent performs internal reasoning based on the multiple pieces of information received, and drives belief adjustment based on the cognitive dissonance caused.

[0018] In some specific embodiments, content strategies include different narrative frameworks and emotional appeals, channel strategies include releases by authoritative media or key opinion leaders, timing strategies include interventions at different points in time after the event occurs, and targeting strategies include broadcasting to the general public or targeted interventions to a predetermined group.

[0019] In some specific implementations, key metrics include the KL divergence between the final belief distribution and the desired goal, the time required to reach stability, and the cost-effectiveness ratio.

[0020] According to a second aspect of the invention, a computer-readable storage medium is provided on which one or more computer programs are stored, which, when executed by a computer processor, implement the method described above.

[0021] According to a third aspect of the present invention, a multi-agent-based population belief evolution and intervention assessment system is proposed, comprising:

[0022] The agent generation module is configured to assign socio-demographic attributes, social network attributes, and psychological cognitive attributes to each agent, and quantify the agent's belief in a predetermined issue into a multi-dimensional continuous vector.

[0023] The environment configuration module configures the construction of a simulated information ecosystem containing multiple information sources, and sets credibility, dissemination intensity, and stance bias attributes for each information source.

[0024] The core of the simulation engine is configured to run simulations and execute agent interaction and belief update rules, which are based on social psychology and communication models.

[0025] An intervention strategy library, including content strategy, channel strategy, timing strategy, and target strategy, is dynamically injected during the simulation process;

[0026] The strategy management module is configured to encapsulate various intervention strategies into pluggable, standardized modules.

[0027] The evaluation and analysis module is configured to run a baseline scenario without intervention and multiple intervention strategy scenarios in a simulation environment. It tracks belief evolution data at the group and individual levels in real time and performs quantitative evaluation and ranking of different intervention strategies based on preset key indicators.

[0028] In some specific embodiments, belief quantization quantifies the agent's belief in a predetermined issue into a multidimensional continuous vector B = [b1, b2, ..., bn], where each dimension represents an aspect of the belief and its value varies continuously in the interval [-1, 1].

[0029] In some specific embodiments, the interaction rules include: the agent is influenced by its social network neighbors, and the influence weight is determined by the closeness of the relationship and the influence of the other party; after receiving information, the degree of belief update of the agent depends on the credibility of the information source, the consistency between the information content and its own existing beliefs, and its own cognitive reflection ability; the agent performs internal reasoning based on the multiple pieces of information received, and drives belief adjustment based on the cognitive dissonance caused.

[0030] In some specific embodiments, content strategies include different narrative frameworks and emotional appeals; channel strategies include releases by authoritative media or key opinion leaders; timing strategies include interventions at different points in time after the event occurs; and target strategies include broadcasting to the general population or targeted interventions to a predetermined group. Key indicators include the KL divergence between the final belief distribution and the expected goal, the time required to reach stability, and the cost-effectiveness ratio.

[0031] This invention proposes a multi-agent-based method and system for evaluating population belief evolution and intervention, which has the following technical advantages: First, the mechanism is clear, starting from the psychological rules of individual behavior at the micro level, and emerging from the bottom up the group-level belief evolution laws, which has stronger interpretability and causal insight compared to black-box machine learning models; Second, it has the ability to predict ahead and test without loss, constructing a "digital sandbox" environment, which can test and optimize various intervention strategies in a low-cost and zero-risk manner before real implementation; In addition, it supports personalized and precise intervention, by meticulously modeling individual differences and social network structures, it can not only evaluate "one-size-fits-all" solutions, but also analyze the effects of precise strategies for specific groups; Finally, it provides a systematic evaluation mechanism, establishing a complete quantitative indicator system to comprehensively evaluate the intervention effect from multiple dimensions, providing strong support for scientific decision-making. Attached Figure Description

[0032] The accompanying drawings are included to provide a further understanding of the embodiments and are incorporated in and constitute a part of this specification. The drawings illustrate embodiments and, together with the description, serve to explain the principles of the invention. Other embodiments and many anticipated advantages of the embodiments will be readily recognized as they become better understood through reference to the following detailed description. Other features, objects, and advantages of this application will become more apparent from reading the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0033] Figure 1 This is a flowchart of a multi-agent-based population belief evolution and intervention assessment method according to an embodiment of this application;

[0034] Figure 2 This is a flowchart of a specific embodiment of the multi-agent-based population belief evolution and intervention assessment method of this application;

[0035] Figure 3 This is a framework diagram of a multi-agent-based population belief evolution and intervention assessment system according to an embodiment of this application;

[0036] Figure 4 This is a schematic diagram comparing the effects of different intervention strategies in a specific embodiment of this application;

[0037] Figure 5 This is a schematic diagram of the structure of a computer system used to implement the electronic device of the present application. Detailed Implementation

[0038] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0039] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0040] Figure 1 A flowchart illustrating a multi-agent-based population belief evolution and intervention assessment method according to an embodiment of this application is shown. Figure 1 As shown, the method includes the following steps:

[0041] S1: Assign each agent a set of multidimensional attributes, including sociodemographic attributes, social network attributes, and psychological cognitive attributes, and quantify the agent's beliefs about a predetermined issue into a multidimensional continuous vector.

[0042] In specific embodiments, each agent (representing an individual) in the simulation environment is assigned a set of multidimensional attributes, including: socio-demographic attributes (such as age, occupation, education level, etc.); social network attributes (such as the agent's position, relationship strength, and influence level in the social network); and psychological cognitive attributes, such as parameters like openness, caution, and conformity based on cognitive psychology theories (such as the five-factor personality model and the Cognitive Reflex Test CRT). Belief quantification quantifies the agent's belief in a predetermined issue into a multidimensional continuous vector B = [b1, b2, ..., bn], where each dimension represents an aspect of the belief, and its value continuously varies within the interval [-1, 1].

[0043] S2: Construct a simulated information ecosystem environment containing multiple information sources, and define the credibility, dissemination intensity, and stance bias attributes of each information source. This includes authoritative media, self-media, social circles, and other information sources. Each information source possesses attributes such as credibility, dissemination intensity, and stance bias. Information is released in the form of "events," and each event includes characteristics such as its content theme, emotional polarity, and strength of argumentation.

[0044] S3: Defines the interaction rules between agents and between agents and environmental information. These interaction rules are based on sociological psychology and communication models.

[0045] In a specific embodiment, the interaction rules include: the agent is influenced by its social network neighbors, and the influence weight is determined by the closeness of the relationship and the influence of the other party; after receiving information, the degree of belief update of the agent depends on the credibility of the information source, the consistency between the information content and its own existing beliefs, and its own cognitive reflection ability; the agent performs internal reasoning based on the multiple information received, and drives belief adjustment based on the cognitive dissonance caused.

[0046] S4: Encapsulate various intervention strategies to be evaluated into pluggable standardized modules. Intervention strategies include content strategies, channel strategies, timing strategies, and target strategies.

[0047] In specific embodiments, content strategies include different narrative frameworks and emotional appeals, channel strategies include releases by authoritative media or key opinion leaders, timing strategies include interventions at different points in time after the event occurs, and target strategies include broadcasting to the general public or targeted interventions to a predetermined group.

[0048] S5: Run a baseline scenario without intervention and multiple intervention strategy scenarios in a simulation environment, track and record belief evolution data at the group and individual levels in real time, and quantitatively evaluate and rank different intervention strategies by comparing key indicators. Key indicators include the KL divergence between the final belief distribution and the expected goal, the time required to reach stability, and the cost-effectiveness ratio.

[0049] Figure 2 A flowchart illustrating a specific embodiment of this application's method for population belief evolution and intervention assessment based on multi-agent systems is shown, as follows: Figure 2 As shown, it includes the following steps:

[0050] Step 1: System initialization, configuring the agent, environment, and initial beliefs by inputting real data. In this stage, the agent's socio-demographic attributes, social network relationships, and psychological cognitive parameters are configured by inputting real data. Simultaneously, the information environment and initial belief states are set, thus laying the foundation for subsequent simulations.

[0051] Step 2: Baseline simulation, running a no-intervention scenario and recording the natural evolution path. The model is run without any intervention strategies to track and record the natural evolution path of population beliefs, serving as a reference for evaluating different intervention strategies.

[0052] Step 3: Strategy Injection. Load the intervention strategy module to be evaluated and execute it at a specific time point. At the predetermined time point of the simulation run, load the intervention strategy module to be evaluated. This strategy can be one or more combinations of content strategy, channel strategy, timing strategy, or target strategy to simulate different intervention measures.

[0053] Step 4: Parallel simulation, running multiple policy scenarios and tracking and recording data throughout the process. In addition to the baseline scenario, multiple different policy scenarios are run simultaneously to track and record the entire process of belief evolution at both the group and individual levels in real time for comparison.

[0054] Step 5: Effectiveness Evaluation. Compare each strategy with the baseline scenario, and output quantitative indicators and a visual report. By comparing each intervention scenario with the baseline scenario, calculate quantitative indicators including the KL divergence between belief distribution and the goal, the time required to reach steady state, the group polarization index, and the cost-benefit ratio, and generate a visual analysis report. If the results do not meet expectations, the strategy can be further optimized based on the indicators.

[0055] Finally, the system outputs decision support results. Based on quantitative assessments and visualization reports, the relatively optimal intervention strategy is determined, and this strategy can be extended to real-world decision-making and management, thereby achieving scientific prediction and effective intervention of population belief evolution under low-risk and low-cost conditions.

[0056] Figure 3 This is a framework diagram of a multi-agent-based population belief evolution and intervention assessment system according to an embodiment of this application, as shown below. Figure 3As shown, the system includes an agent generation module 301, used to construct AI agents with multi-dimensional attributes based on real-world data. Real-world data includes demographic information, social media data, and survey data. This module can assign socio-demographic attributes (such as age, occupation, and education level), social relationship attributes (such as network location, relationship strength, and influence level), and psychological cognitive attributes (such as personality traits and cognitive bias parameters) to each agent, thereby forming a heterogeneous group of agents. An environment configuration module 302 is used to define the simulated information ecosystem, constructing a multi-source information environment including authoritative media, self-media, and social circles, and setting attributes such as the credibility, dissemination intensity, and stance bias of information sources. Simultaneously, this module is also responsible for establishing the social network topology between agents, ensuring the authenticity of group interactions. A simulation engine core 303, as the system's operational core, drives the interaction between agents and between agents and environmental information. This module defines interaction rules based on social psychology and communication models, enabling agents to dynamically update their belief vectors after receiving information based on information credibility, matching with existing beliefs, and their own cognitive reflection capabilities. The simulation engine core supports long-step dynamic simulations, capable of simulating the evolution of group beliefs from an initial state to a steady state. The intervention strategy library 304 stores various predefined intervention strategy modules. Intervention strategies include at least content strategies, channel strategies, timing strategies, and target strategies. Different strategies can be invoked individually or in combination to simulate diverse intervention measures in real-world scenarios. The strategy management module 305 is used to invoke and inject intervention strategies and dynamically manage them during the simulation process. This module allows strategies to be injected into the simulation engine in a pluggable manner and supports implementation at different time points and within different group scopes, thus flexibly controlling the intervention execution process. Simultaneously, the strategy management module is also responsible for feedback optimization of evaluation results, forming a closed loop of strategy iterative improvement. The evaluation and analysis module 306 is used to quantitatively analyze and visualize the results after the simulation. This module can collect belief evolution data at the group and individual levels in real time, calculate multi-dimensional evaluation indicators including KL divergence, polarization index, belief stabilization time, and cost-effectiveness ratio, and output an evaluation report. The report not only includes numerical indicators, but also provides curve comparisons and visual evolution results to intuitively demonstrate the differences in the effects of different intervention strategies.

[0057] Through the synergistic effect of the above modules, the system of this application can perform high-fidelity simulation of the evolution of people's beliefs in a virtual environment, and realize non-destructive testing and quantitative evaluation of various intervention strategies, thereby providing scientific support for decision-making in the real world.

[0058] Figure 4 This is a schematic diagram illustrating the comparative evaluation of the effects of different intervention strategies in a specific embodiment of this application, as shown below. Figure 4As shown, the horizontal axis represents the simulation time (days), and the vertical axis represents the average group belief value, ranging from -1 to 1. The curves of different colors represent the dynamic evolution trajectory of the average group belief over time under different intervention strategies. It can be seen that beliefs naturally slide towards the negative (-0.95) without external intervention (baseline scenario). Strategy A (authoritative media refutation) showed the best results. This strategy, through the release of strongly argued information via highly credible official channels, was significant and robust, successfully guiding group beliefs towards a positive goal (+0.92). Strategy B (key opinion leader guidance) was less effective. Although it utilized influential nodes (KOLs), their credibility or argumentation methods may not have matched the target group, failing to effectively reverse beliefs and even leading to a rebound later. Strategy C (precise penetration into social networks) had limited and fluctuating effects. This strategy achieved weak early results through small-scale, targeted penetration into social networks, but its insufficient influence made it difficult to change the overall trend, and the effect subsequently diminished. This figure visually illustrates the differences in how different intervention strategies influence the evolution of group beliefs. By comparing the final convergence value, stabilization time, and evolutionary trend of each curve, the effectiveness, efficiency, and robustness of the strategies can be quantitatively evaluated, thus providing a scientific basis for subsequent strategy optimization and decision support.

[0059] The following is for reference. Figure 5 It shows a schematic diagram of the structure of a computer system suitable for implementing the electronic device of the present application. Figure 5 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0060] like Figure 5 As shown, the computer system includes a central processing unit (CPU) 501, which can perform various appropriate actions and processes based on programs stored in read-only memory (ROM) 502 or programs loaded from storage section 508 into random access memory (RAM) 503. RAM 503 also stores various programs and data required for the operation of system 500. CPU 501, ROM 502, and RAM 503 are interconnected via bus 504. Input / output (I / O) interface 505 is also connected to bus 504.

[0061] The following components are connected to I / O interface 505: an input section 506 including a keyboard, mouse, etc.; an output section 507 including a liquid crystal display (LCD) and speakers, etc.; a storage section 508 including a hard disk, etc.; and a communication section 509 including a network interface card such as a LAN card and a modem, etc. The communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to I / O interface 505 as needed. A removable medium 511, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 510 as needed so that computer programs read from it can be installed into storage section 508 as needed.

[0062] Specifically, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable storage medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 509, and / or installed from removable medium 511. When the computer program is executed by central processing unit (CPU) 501, it performs the functions defined in the methods of this application. It should be noted that the computer-readable storage medium of this application can be a computer-readable signal medium or a computer-readable storage medium or any combination thereof. The computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can also be any computer-readable storage medium other than a computer-readable storage medium that can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. Program code contained on a computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0063] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages—such as Java and C++—and conventional procedural programming languages—such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0064] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0065] The modules described in the embodiments of this application can be implemented in software or in hardware.

[0066] In another aspect, this application also provides a computer-readable storage medium, which may be included in the electronic device described in the above embodiments; or it may exist independently and not assembled into the electronic device. The aforementioned computer-readable storage medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to: assign a set of multidimensional attributes to each agent, including socio-demographic attributes, social network attributes, and psychological cognitive attributes, and quantify the agent's beliefs about a predetermined issue into a multidimensional continuous vector; construct a simulated information ecosystem environment containing multiple information sources, and define the credibility, dissemination intensity, and stance bias attributes of each information source; define interaction rules between agents and between agents and environmental information, the interaction rules being based on sociological psychology and communication models; encapsulate multiple intervention strategies to be evaluated into pluggable standardized modules, the intervention strategies including content strategies, channel strategies, timing strategies, and target strategies; run a baseline scenario without intervention and multiple intervention strategy scenarios in a simulation environment, track and record belief evolution data at the group and individual levels in real time, and quantitatively evaluate and rank different intervention strategies by comparing key indicators.

[0067] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described inventive concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this application.

Claims

1. A method for assessing population belief evolution and intervention based on multi-agent systems, characterized in that, include: S1: Assign each agent a set of multidimensional attributes, including sociodemographic attributes, social network attributes, and psychological cognitive attributes, and quantify the agent's belief in a predetermined issue into a multidimensional continuous vector; S2: Construct a simulated information ecosystem environment containing multiple information sources, and define the credibility, dissemination intensity, and stance bias attributes of each information source; S3: Define the interaction rules between the intelligent agents and between the intelligent agents and environmental information, the interaction rules being based on sociological psychology and communication models; S4: Encapsulate multiple intervention strategies to be evaluated into pluggable standardized modules, including content strategies, channel strategies, timing strategies, and target strategies; S5: Run baseline scenarios without intervention and multiple intervention strategy scenarios in a simulation environment, track and record belief evolution data at the group and individual levels in real time, and quantitatively evaluate and rank different intervention strategies by comparing key indicators.

2. The method for population belief evolution and intervention assessment based on multi-agent systems according to claim 1, characterized in that, The belief quantization involves quantifying the agent's belief in the predetermined issue into a multidimensional continuous vector B = [b1, b2, ..., bn], where each dimension represents an aspect of the belief, and its value varies continuously within the interval [-1, 1].

3. The method for population belief evolution and intervention assessment based on multi-agent systems according to claim 1, characterized in that, The interaction rules include: the agent is influenced by its social network neighbors, and the influence weight is determined by the closeness of the relationship and the influence of the other party; after receiving information, the degree of belief update of the agent depends on the credibility of the information source, the consistency between the information content and its own existing beliefs, and its own cognitive reflection ability; the agent performs internal reasoning based on the multiple pieces of information received, and drives belief adjustment based on the cognitive dissonance caused.

4. The method for population belief evolution and intervention assessment based on multi-agent systems according to claim 1, characterized in that, The content strategy includes different narrative frameworks and emotional appeals; the channel strategy includes releases by authoritative media or key opinion leaders; the timing strategy includes interventions at different points in time after the event occurs; and the targeting strategy includes broadcasting to the general public or targeted interventions to a predetermined group.

5. The method for population belief evolution and intervention assessment based on multi-agent systems according to claim 1, characterized in that, The key indicators include the KL divergence between the final belief distribution and the expected goal, the time required to reach stability, and the cost-benefit ratio.

6. A computer-readable storage medium having one or more computer programs stored thereon, characterized in that, When the one or more computer programs are executed by a computer processor, they perform the method according to any one of claims 1-5.

7. A multi-agent-based population belief evolution and intervention assessment system, characterized in that, include: The agent generation module is configured to assign socio-demographic attributes, social network attributes, and psychological cognitive attributes to each agent, and quantify the agent's belief in a predetermined issue into a multi-dimensional continuous vector. The environment configuration module configures the construction of a simulated information ecosystem containing multiple information sources, and sets credibility, dissemination intensity, and stance bias attributes for each information source. The core of the simulation engine is configured to run simulations and execute intelligent agent interaction and belief update interaction rules, which are based on social psychology and communication models. An intervention strategy library, including content strategy, channel strategy, timing strategy, and target strategy, is dynamically injected during the simulation process; The strategy management module is configured to encapsulate various intervention strategies into pluggable, standardized modules. The evaluation and analysis module is configured to run a baseline scenario without intervention and multiple intervention strategy scenarios in a simulation environment. It tracks belief evolution data at the group and individual levels in real time and performs quantitative evaluation and ranking of different intervention strategies based on preset key indicators.

8. A multi-agent-based population belief evolution and intervention assessment system according to claim 7, characterized in that, The belief quantization involves quantifying the agent's belief in the predetermined issue into a multidimensional continuous vector B = [b1, b2, ..., bn], where each dimension represents an aspect of the belief, and its value varies continuously within the interval [-1, 1].

9. A multi-agent-based population belief evolution and intervention assessment system according to claim 7, characterized in that, The interaction rules include: the agent is influenced by its social network neighbors, and the influence weight is determined by the closeness of the relationship and the influence of the other party; after receiving information, the degree of belief update of the agent depends on the credibility of the information source, the consistency between the information content and its own existing beliefs, and its own cognitive reflection ability; the agent performs internal reasoning based on the multiple pieces of information received, and drives belief adjustment based on the cognitive dissonance caused.

10. A multi-agent-based population belief evolution and intervention assessment system according to claim 7, characterized in that, The content strategy includes different narrative frameworks and emotional appeals; the channel strategy includes releases by authoritative media or key opinion leaders; the timing strategy includes interventions at different points in time after the event occurs; the target strategy includes broadcasting to the general population or targeted interventions to a predetermined group; the key indicators include the KL divergence between the final belief distribution and the expected goal, the time required to reach stability, and the cost-effectiveness ratio.

Images