基于Multiagent的社交网络舆情传播模型  

Multiagent Public Sentiment Propagation Model of Social Network

作　　者：董田田 李军 王继荣[2] DONG Tiantian;LI Jun;WANG Jirong(College of Computer Science&Technology,Qingdao University,Qingdao 266071,China;College of Mechanical and Electrical Engineering,Qingdao University,Qingdao 266071,China)

机构地区：[1]青岛大学计算机科学技术学院,山东青岛266071 [2]青岛大学机电工程学院,山东青岛266071

出　　处：《青岛大学学报（工程技术版）》2021年第2期28-35,共8页Journal of Qingdao University(Engineering & Technology Edition)

基　　金：山东省自然科学基金资助项目(ZR2019MF050)。

摘　　要：为分析与预测社交网络中舆情传播过程和演化趋势,本文基于Multiagent分布式技术,建立了社交网络舆情传播模型,研究社交网络舆情传播方式,并根据Agent的行为和状态特性,对Agent进行分类,分析Agent的属性特征,通过剖析Agent个体之间的交互微观行为,依据信息价值时效性、信息源覆盖率和信息关注度,对转发行为的影响进行讨论,给出了Agent的决策规则,以网络结构方式体现Agent之间的内在联系。为验证模型的有效性,基于Agent的仿真平台,采用Netlogo 6.0.4进行模拟仿真分析。仿真结果表明,舆情信息的传播与信息价值呈正相关,信息价值越大,越有利于舆情信息的传播,且信息源的覆盖率与转发Agent数和中止Agent数成正相关,信息关注度对舆情传播影响甚大,说明该模型能准确分析和预测网络中舆情传播的演化趋势。该研究对相关人员了解社交网络中舆情的传播具有重要意义。In order to analyze and predict the process and evolution of public opinion communication in social networks,this paper establishes a social network public opinion communication model based on Multiagent distributed technology.It studies the way of spreading public opinion on social networks,and classifies Agents according to their behavior and status characteristics,and analyzes the attributes of Agents.It analyzes the interaction micro-behavior between individual Agents.And based on the timeliness of information value,coverage of information sources and Information attention,the influence of forwarding behavior is discussed,the decision-making rules of the agent are given,and the internal connection between the agents is reflected in the network structure.In order to verify the validity of the model,based on the Agent simulation platform,Netlogo 6.0.4 is used for simulation analysis.The simulation results show that the value of information promotes the dissemination of public opinion information.The greater the value of the information,the more conducive to the dissemination of public opinion information.The coverage rate of the information source is positively correlated with the number of forwarding agents and the number of suspended agents.The dissemination of public opinion has a great influence,which shows that the model can accurately analyze and predict the evolution trend of public opinion dissemination in the network.This research is of great significance for relevant personnel to understand the spread of public opinion in social networks.

关 键 词：MULTIAGENT 社交网络 舆情话题 传播模型 价值时效性 信息源覆盖率 

分 类 号：TP393.034[自动化与计算机技术—计算机应用技术]