基于自适应策略的网络舆论演化
发布时间：2018-10-12 19:05
【摘要】：社交网络上的舆论演化过程是现今国内外信息传播、复杂系统、行为分析等领域的一个研究热点。针对现有研究常受制于固定的网民观点交换、更新策略,而难以还原真实网络舆论演化过程的问题,提出了一种基于自适应策略的网络舆论演化模型。首先,选取BA(Barabási和Albert)无标度网络和WS(Watts和Strogatz)小世界网络作为网络模型。然后,结合Sznajd模型与多数准则模型的思想,设计了两种观点更新策略。最后,为保证网民所持观点在全局网络中占据优势,设置了自适应的观点更新策略选取规则。仿真实验结果表明,随着舆论演化时间的增长,网民选取固定策略的概率逐渐趋近于1;最终选择说服邻居策略的网民数量与选择多数观点策略的网民数量基本相等,这种变化趋势不依赖于初始观点分布和网络拓扑结构。在舆论演化过程中,节点中心度大于100的网民均有固定的观点更新策略,而中心度小于10的网民常动态改变策略。仿真实验结果显示,WS网络和BA网络中持少数观点的网民数量会随着时间增长而减少,但WS网络中的少数观点网民数量减少幅度更大。
[Abstract]:The evolution of public opinion on social networks is a hot topic in the fields of information dissemination, complex systems, behavior analysis and so on. In order to solve the problem that the current research is often restricted by the fixed exchange of views and updated strategies of Internet users, it is difficult to restore the evolution process of real network public opinion. A model of network public opinion evolution based on adaptive strategy is proposed in this paper. Firstly, BA (Barab 谩 si and Albert) scale-free networks and WS (Watts and Strogatz) small-world networks are selected as network models. Then, combining the idea of Sznajd model and majority criterion model, two strategies of view updating are designed. Finally, in order to ensure that the view of Internet users occupies the advantage in the global network, an adaptive selection rule of view updating strategy is set up. The simulation results show that with the development of public opinion, the probability of Internet users choosing a fixed strategy approaches to 1; the number of netizens who choose the strategy of persuading neighbors is equal to the number of netizens who choose the strategy of majority opinion. This trend does not depend on the initial viewpoint distribution and network topology. In the process of public opinion evolution, netizens with node centrality greater than 100 have a fixed strategy to update their views, while those with center degree less than 10 often change their strategies dynamically. The simulation results show that the number of minority netizens in WS network and BA network will decrease with time, but the number of minority Internet users in WS network will decrease even more.
【作者单位】： 西华大学计算机与软件学院;
【基金】：国家自然科学基金资助项目(61602389) 四川省教育厅科研项目(15ZB0133) 西华大学自然科学重点基金项目(z1422617);西华大学省部级学科平台开放课题(szjj2015-58) 互联网自然语言智能处理四川省高等学校重点实验室资助项目(INLP201501)
【分类号】：O157.5
[Abstract]:The evolution of public opinion on social networks is a hot topic in the fields of information dissemination, complex systems, behavior analysis and so on. In order to solve the problem that the current research is often restricted by the fixed exchange of views and updated strategies of Internet users, it is difficult to restore the evolution process of real network public opinion. A model of network public opinion evolution based on adaptive strategy is proposed in this paper. Firstly, BA (Barab 谩 si and Albert) scale-free networks and WS (Watts and Strogatz) small-world networks are selected as network models. Then, combining the idea of Sznajd model and majority criterion model, two strategies of view updating are designed. Finally, in order to ensure that the view of Internet users occupies the advantage in the global network, an adaptive selection rule of view updating strategy is set up. The simulation results show that with the development of public opinion, the probability of Internet users choosing a fixed strategy approaches to 1; the number of netizens who choose the strategy of persuading neighbors is equal to the number of netizens who choose the strategy of majority opinion. This trend does not depend on the initial viewpoint distribution and network topology. In the process of public opinion evolution, netizens with node centrality greater than 100 have a fixed strategy to update their views, while those with center degree less than 10 often change their strategies dynamically. The simulation results show that the number of minority netizens in WS network and BA network will decrease with time, but the number of minority Internet users in WS network will decrease even more.
【作者单位】： 西华大学计算机与软件学院;
【基金】：国家自然科学基金资助项目(61602389) 四川省教育厅科研项目(15ZB0133) 西华大学自然科学重点基金项目(z1422617);西华大学省部级学科平台开放课题(szjj2015-58) 互联网自然语言智能处理四川省高等学校重点实验室资助项目(INLP201501)
【分类号】：O157.5
【相似文献】
相关期刊论文 前3条
1 谭伟伟;;网格自适应策略在高升力构型计算中的应用[J];航空计算技术;2010年06期
2 严心池,安伟光,赵维涛;自适应免疫遗传算法[J];应用力学学报;2005年03期
3 ;[J];;年期
相关会议论文 前3条
1 徐慧;肖德宝;刘会芬;徐艳;;效用计算环境下基于自适应策略的资源分配模型的研究[A];第一届中国高校通信类院系学术研讨会论文集[C];2007年
2 蔡畅;戚文军;周楠;;基于自适应策略的铝型材在线淬火控制系统设计[A];2013广东材料发展论坛——战略性新兴产业发展与新材料科技创新研讨会论文摘要集[C];2013年
3 张翔;李纲;熊伟清;;修正AHP中判断矩阵一致性的粒子群优化算法[A];第二十九届中国控制会议论文集[C];2010年
相关博士学位论文 前2条
1 葛亮;面向环境状态的ASBS系统自适应策略评价方法研究[D];东北大学;2014年
2 赵春蕾;端信息跳变系统自适应策略研究[D];南开大学;2012年
相关硕士学位论文 前10条
1 常德兴;基于自适应策略的实时目标跟踪方法[D];西南交通大学;2016年
2 解高纯;云编程环境下的程序运行性能自适应模拟方法研究[D];北京工业大学;2016年
3 聂素丽;重复性公共品博弈中的合作演化[D];西安电子科技大学;2015年
4 苏世雄;动态软件体系结构模型及自适应策略的研究[D];兰州交通大学;2016年
5 龚e，
本文编号：2267266
本文链接：https://www.wllwen.com/kejilunwen/yysx/2267266.html
