摘要
为提高高校网络舆情的可识别性和预警实时性,提高网络舆情热点分析的准确性,论文设计了一个高校网络舆情热点发现模型.包括网络舆情信息采集、预处理、中文分词、特征选择、文本分词和聚类分析.考虑到网络舆情的不确定性和模糊性,提出了一种基于信息熵和密度改进的K-Means聚类算法的网络舆情相似度分析方法,此方法可以对网络热点和危机事件进行聚类和识别.实验结果表明,该方法能够快速获得网络舆情,具有较高的聚类准确率,证明了论文提出的模型的可行性与有效性,可为高校网络舆情监测和识别提供重要的技术支持.
In order to improve the timeliness of identification and early warning of network public opinion in colleges,and improve the accuracy of online hotspot analysis,a hotspot discovery model for internet public opinion in colleges was designed.Including internet public opinion information collection,preprocessing,Chinese word segmentation,feature selection,text segmentation and cluster analysis.Considering the uncertainty and fuzziness of the Internet public opinion,a method for analyzing the similarity of the Internet based on the information entropy and density improved K-Means clustering algorithm is proposed,which can solve the network hotspot and crisis events.Events are clustered and identified.The experimental results show that the method can quickly obtain the Internet public opinion,has a high accuracy of clustering,and proves the feasibility and effectiveness of the model.It provides important technical support for the monitoring and identification of public opinion in colleges.
作者
陈艳红
向军
刘嵩
CHEN Yanhong;XIANG Jun;LIU Song(School of Information Engineering,Hubei University for Nationalities,Enshi 445000,China)
出处
《湖北民族学院学报（自然科学版）》
CAS
2018年第4期442-447,共6页
Journal of Hubei Minzu University(Natural Science Edition)
基金
国家自然科学基金项目(61362016)
文化部科技提升计划项目(201307)
湖北省自然科学基金项目(2009CBD069)
关键词
信息熵
K-MEANS
网络舆情
information entropy
K-Means
internet public opinion
