基于自动分类的网络舆情监测方法研究

Research on Opinion Monitoring Method Based on Automatic Text Classification

当前互联网快速发展,网络社会与现实社会逐渐同步,网上网下事件的关联性提高,网络舆情也越来越能够及时反映现实社会中发生的事情。因此,网络舆情监测不仅能够了解民意,为相关决策部门制定方案提供参考,而且能够通过大数据分析,对突发事件进行及时预警。以互联网上微博、贴吧、论坛、新闻评论等信息作为对象,以实用性为原则,研究一种基于文本自动分类的网络舆情监测方法。该方法通过网络爬虫抓取互联网上的信息,并采用基于KNN算法的文本自动分类方式完成网络舆情自动分类,最后通过实验验证了该方法的实用性。

With the rapid development of the Internet, the cyber society and the real society are gradually synchronized, and incidents on the Internet have become more and more related to incidents offline. Public opinions online are also be- coming more capable of reflecting social events in time. Therefore, network public opinion monitoring not only helps to in- form public opinions, providing references for the decision--making departments in concern, but also gives timely warning of unexpected events thanks to large data analysis. This study, focusing on Internet information such as micro blog, post bar, forum and news comments, explores, on the principle of applicability, an Internet public opinion monitoring method which is based on automatic text classification. A web crawler is applied to collect information on the Internet, and the KNN algorithm is used to conduct the automatic classification of online public opinions. The applicability of this method is verified by experiments at the end of the current study.