基于多维度的公安舆情分析模型构建

目的/意义]旨在为有关部门及时分析和应对舆情提供参考。方法/过程]通过爬虫搜集媒体上的舆情数据,计算舆情热度并划分舆情演化阶段,利用TextRank算法提取不同舆情阶段的关键词,采用Python语言的Snownlp库进行文本的情感分析,构建公安舆情分析模型,并通过“六安事件”对该模型进行验证。结果/结论]该模型从热度、关键词与情感三个维度较好地对公安舆情数据进行分析和挖掘,有助于及时有效地处理社交媒体上的公安舆情数据,为公安舆情分析提供新的途径。

Construction of Public Security Public Opinion Analysis Model Based on Multi-Dimensions

Purpose/significance]The paper is to provide references for the relevant departments to timely analyze and deal with the public opinion.Method/process]The paper collects the public opinion data on the media by crawler,calculates the public opinion heat and divides the public opinion evolution stage,uses the TextRank algorithm to extract keywords of the different public opinion stages,uses the Python language’s Snownlp library to analyze the emotion of texts,constructs the public security public opinion analysis model,which is verified by“Liu’an event”.Result/conclusion]The model can analyze and mine public security public opinion data from three dimensions of heat,keywords and emotion,which helps to deal with public opinion data in social media in time and effectively,and provides a new way for public opinion analysis.