BBS突发舆情分析及基于小世界网络的预测模型  

Analysis of emergent BBS sentiment and its prediction model based on small world network

作　　者：吴渝[1] 杨涛[1] 肖开洲[1] 

机构地区：[1]重庆邮电大学网络智能研究所,重庆400065

出　　处：《重庆邮电大学学报（自然科学版）》2010年第3期350-354,共5页Journal of Chongqing University of Posts and Telecommunications(Natural Science Edition)

基　　金：973计划前期研究专项课题(2008CB317111);国家自然科学基金(60873079);新世纪优秀人才支持计划;重庆市重点自然科学基金(2009BA2089;2008BB2241)~~

摘　　要：BBS是网络舆情产生和传播的主要场所之一,由于研究手段的匮乏,预测和引导BBS舆情的研究工作仍处于探索阶段。考虑到网络舆情具有突现性,以突现计算的观点将发帖人抽象成Agent。通过设定Agent参数和交互规则,利用小世界网络建立了BBS舆情预测模型。仿真结果表明,在日增回帖数和个人发言数比例这2个重要统计指标上与真实情况同时吻合,说明了模型的合理性和有效性。BBS has been one of the major places of the generation and propagation of public sentiment.Due to the deficiency of methods,research on prediction and control of public sentiment is still in its infancy.Since emergency is one of the features of network sentiment,the author of a poster is Abstracted as an agent.By setting the parameter and interaction rules of all the agents,a prediction model of the public sentiment based on small world network was built.Simulation results show that this model is consistent with the real environment on two key statistics——reply addition per day and personal speech proportion.So it is effective and efficient.

关 键 词：BBS舆情 突现计算 小世界网络 预测模型 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]