基于复杂网络的协同舆情演化模型研究  

RESEARCH ON SYNERGISTIC OPINION EVOLUTION MODEL BASED ON COMPLEX NETWORKS

作　　者：孙士保[1] 张亚楠[1] 张京山 章冲[1] 

机构地区：[1]河南科技大学信息工程学院

出　　处：《计算机应用与软件》2017年第6期52-56,68,共6页Computer Applications and Software

基　　金：国家自然科学基金项目(51474095);河南省重点攻关项目(152102210277);河南省产学研合作项目(152107000027);河南省高校科技创新团队支持计划项目(17IRTSTHN010);河南科技大学科技创新团队项目(2015XTD011);河南科技大学重大产学研合作培育基金项目(2015ZDCXY03)

摘　　要：针对舆情传播中人际关系网络的复杂性和个体邻域的异质性,建立可变聚类系数的无标度网络,在此基础上构建了协同舆情演化模型。通过不同参数下的仿真实验,分别讨论了序参量和网络结构特征对系统磁化率的影响。实验结果表明,序参量更能体现舆情系统的协同效应,比较符合网络舆情演化的实际情况。此外,当网络平均度较小时,提高聚类系数可以有效抑制舆情的传播;当平均度超过临界值时,将促使系统迅速收敛于极端态度,此时聚类系数不再起关键作用。研究结果有助于理解舆情的演化规律,为导控舆情发展提供了理论参考。Based on the complexity of interpersonal relationship network and the heterogeneity of individualneighborhood, a scale-free network with variable clustering coefficient is established. After that, a collaborative public opinion evolution model is constructed. The influence of the order parameter and network structure on the magnetic susceptibility of the system is discussed through simulation experiments under different parameters. The experimental results show that the order parameter can reflect the synergy effect of the public opinion system, which is more consistent with the actual situation of network public opinion evolution. In addition, when the network average is small, increasing the clustering coefficient can effectively suppress the spread of public opinion ; when the average degree exceeds the critical value, will prompt the system to converge to the extreme attitude quickly, then clustering coefficient no longer plays a key role. The results can help to understand the law of evolution of public opinion and provide a theoretical reference for controlling the development of public opinion.

关 键 词：舆情演化 复杂网络 协同效应 平均度 聚类系数 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]