微博作为目前主流的互联网社交媒体,群众可在其上随时随地发布“河湖长制”相关的事件信息。通过微博对感兴趣事件进行提取与分类可以有效提高“河湖长制”管理过程中问题事件的发现与解决效率,并针对群众对河湖岸线管理状态评价进行有效的监测与分析。随着“互联网+”技术在河湖长制中发挥巨大作用,对微博社交大数据在“河湖长制”管理工作中的应用进行了探讨,基于互联网社交媒体中含有的大量关于“河湖长制”事件的文字描述信息,提出了微博社交文本信息挖掘系统框架,采用互联网爬取技术和语义分析技术抓取关于“河湖长制”事件的新闻和公众信息,并对事件进行识别,提取时间、位置和事件类型等标签,最后进行数据挖掘分析情感倾向。以2018年6月至2020年12月微博上珠江河网区数据为实例,情感分析模型在测试集上的事件分类准确率为88.6%,证明该模型具有一定的可用性。该舆情分析系统可极大提高了“河湖长制”管理效率。

As the current mainstream Internet social media, Weibo allows its users to post information about events involved with the “River Chief System” anytime and anywhere on it. Extracting and classifying events improves the efficiency of discovering and solving problematic events in the management of river chief system, and also helps monitor and analyze the evaluation of the management status of river and lake shorelines for the people. The “Internet +” technology has been playing a tremendous role in the management of river chief system. In the present paper, the application of Weibo’s big data in the management of river chief system is discussed at first. A framework of text information mining system for Weibo is proposed based on the large amount of text description contained in social media. News and public information about events involved with river chief system can be captured by using Internet crawling technology and semantic recognition analysis technology via this framework. Tags of the events such as time, location, and type are identified and extracted, and finally people’s emotional tendencies are analyzed through data mining. With the data of Pearl River network from June 2018 to December 2020 on Weibo as a case study, the correctness of the model in training set reaches 88.6%, implying the feasibility of the model. The public opinion analysis system can help improve the management efficiency of the river chief system.