基于K-中心轮换算法的湖南省移动互朕网网络舆情服务系统的二次开发  

The Second Development of Online Public Opinion Service System of Hunan Mobile Internet Based on K-center Rotation Algorithm

作　　者：潘建明 赵晓军[2] 

机构地区：[1]湖南交通工程学院,湖南衡阳421009 [2]湖南工学院经济与管理学院,湖南衡阳421008

出　　处：《湖南工业职业技术学院学报》2016年第1期5-8,共4页Journal of Hunan Industry Polytechnic

摘　　要：移动互联网舆情传播具有用户数量庞大,发展速度快,信息交流和传播具有泛在性、永久在线性、交流的碎片化和舆论的趋同性的特点,舆情安全变得更加复杂。舆情监控服务至少要实现大数据采集、大数据挖掘和信息源定位三个功能。当前的网络舆情服务系统数据挖掘效率低、聚类分析功能弱,利用k-中心轮换算法可以避免出现局部最优,提高聚类分析的精准度,但仍存在需要事先输入簇数、伸缩性较差等缺点。Mobile internet public opinion has a huge number of users and the rapid development. The information exchange and dissemination has features of being extensive, being permanent, the fragment communicating, and the convergence of public opinion,which make the public opinion security more complex. Public opinion monitoring service at least realizes three functions of the big data acquisition, the data mining and the information source location. As the mining efficiency of the current network public opinion service system data is low, and the clustering analysis function is weak, the use of k- center rotation algorithm can avoid local optimization and improve the accuracy of clustering analysis, but there are some weaknesses： inputing the number of clusters in advance, the poor scalability, etc.

关 键 词：移动互联网 网络舆情 数据挖掘 

分 类 号：TP393.02[自动化与计算机技术—计算机应用技术]