Computer Integrated Manufacturing System ›› 2021, Vol. 27 ›› Issue (1): 248-258.DOI: 10.13196/j.cims.2021.01.023

Previous Articles     Next Articles

Interaction and production model of public opinion based on Agent's emotional persuasion

  

  • Online:2021-01-31 Published:2021-01-31
  • Supported by:
    Project supported by the National Science Foundation,China(No.71972177),and the Fundamental Research Funds for Central Universities,China(No.2020YJSGL04).

基于Agent的情感劝说的舆情交互及产生模型

伍京华,陈虹羽,汪文生+   

  1. 中国矿业大学(北京)管理学院
  • 基金资助:
    国家自然科学基金资助项目(71972177);中央高校基本科研业务费专项资金资助项目(2020YJSGL04)。

Abstract: In view of insufficient quantitative research on public opinion based on agent's emotional persuasion,the interaction and production model of public opinion was constructed.According to the theory of social influence,the Agent's social influence and emotion were identified as the two most important factors that affect the interaction and production of public opinion.The evaluation value of persuasion proposal was obtained with the proposed algorithm for evaluating an emotional persuasion proposal based on Agent's prospect value.Meanwhile,the updating algorithm of public opinion based on Hegselmann-Krause model was proposed by considering Agent's social influence,emotion and personality.Through the numerical example of coal procurement,the simulation comparison and parameter sensitivity analysis of the trust threshold,the relationship threshold and the persuasiveness were used to compare with Hegselmann-Krause model,and the proposed model was more in line with the characteristics of the interaction and production of public opinion based on emotional persuasion.The intelligent advantage of agent was better utilized,and the validity of the model was verified.

Key words: Agent, emotional persuasion, public opinion, interaction of production model, Hegselmann-Krause model

摘要: 针对基于Agent的情感劝说的舆情量化研究不足的问题,构建了基于Agent的情感劝说的舆情交互及产生模型。首先结合社会影响理论,识别了Agent的社会影响力和情感是影响舆情交互及产生的两个最重要因素;其次,通过提出基于Agent的前景值的情感劝说提议评价算法来求取劝说提议的评价值,并考虑Agent的社会影响力、情感及个性,提出了基于Hegselmann-Krause模型的情感劝说舆情更新算法;最后,通过煤炭采购算例、仿真比较及信任阈值、关系度阈值和说服力的参数敏感性分析,说明所构建模型比Hegselmann-Krause模型更符合情感劝说环境下舆情交互及产生特征,更好发挥了Agent的智能化优势,从而验证了所提模型的有效性。

关键词: 智能体, 情感劝说, 舆情, 交互及产生模型, Hegselmann-Krause模型

CLC Number: