基于观点场模型的微博评论观点演化趋势预测方法  

An Opinion-Field-Model-Based Method for Prediction of the Evolution Trends of Opinions of Microblog Comments

作　　者：吴诗贤[1] 张必兰[2] Wu Shixian;Zhang Bilan(Computer Science and Information Engineering College,Chongqing Technology and Business University,Chongqing,400067,China;Library,Chongqing Technology and Business University,Chongqing 400067,China)

机构地区：[1]重庆工商大学计算机科学与信息工程学院,重庆400067 [2]重庆工商大学图书馆,重庆400067

出　　处：《现代情报》2018年第9期74-78,共5页Journal of Modern Information

基　　金：教育部人文社会科学研究项目"基于观点势场演化的舆情极化预测模型研究"(项目编号:13YJCZH197)

摘　　要：[目的 /意义]探讨网络舆情事件中群体观点的演变规律,构建有效预测其演变趋势的方法。[方法 /过程]参考物理学中场的思想和信息科学中数据场的方法,引入观点场概念,提出了一种基于观点势的观点潜在影响力评估模型;然后将该模型运用到微博评论的群体观点演化分析中,建立了微博评论的观点趋势预测方法。该方法的基本思想是以当前评论的观点势分布来预测未来评论的观点分布,在观点势计算时,以既有评论的排序值代表新的信息受众所处的参考场点与观点场中既有评论之间的距离。[结果 /结论]通过实际的微博舆情事件数据实验表明,该网络舆情群体观点趋势预测模型能较好地评估已发表的显性观点对后来网民观点形成的影响力,具有较高的网络舆情观点趋势预测准确率。[Purpose/Significance]To explore the formation and evolution of group opinion trends in cyber public opinion events,and to construct a corresponding method for efficient prediction of the evolution trends.[Method/Process]Based on the analysis of the generation and evolution of viewpoints in network public opinion events,the concept of opinion field was introduced with reference to the idea of physical fields and data fields in information science.An evaluation model for the impact of opinion potential was thus proposed.The model was applied to the evolution analysis of the group opinions in microblog comment groups to establish a method for protection of such evolution trends.The basic idea of this method was to predict the viewpoints distribution of future comments&opinions based on the distribution of the current opinion potential,the quantification of which was based on the rankings of current comments that represented the distance between the reference field point(in which the new information audience was located)and the existing comments in the opinion field.[Result/Conclusion]The actual data experiments on microblog public opinion events showed that the forecasting model of the evolution trend of network public opinion&group opinion could better evaluate the influence of the published explicit opinion on the forthcoming formation of the netizens opinions,and predict more accurately the corresponding evolution trends.

关 键 词：群体极化 观点场 观点势 影响力 评估模型 微博 评论 演变规律 演化趋势 网络舆情 舆论趋势 预测 

分 类 号：G202[文化科学—传播学]