摘要
网络评论观点挖掘的实质就是文本的情感倾向性分类,随着互联网技术的发展,人们更加倾向于在网络上发表自己的意见、态度、观点等,而观点的挖掘能够帮助个人、企业或者政府机构做出相应的决策;在过去的研究中,研究人员主要是在文档级、语句级以及属性级这三个层次进行观点的挖掘,也有一些优秀的综述总结过这几个层次的研究现状,但是还没有整体介绍这三个方面的综述。介绍这三个层次的观点挖掘研究现状:首先,介绍每个层观点挖掘的定义;然后,介绍观点挖掘使用的方法;最后,比较各种方法的优缺点,并总结使用比较多的情感词典。
The essence of online commentary mining is the classification of sentiment orientation of texts. With the development of Internet technology, people are more inclined to express their opinions, attitudes, opinions, etc. on the Internet, and the mining of opinion can help individuals, enterprises or government to make corresponding decisions;in the past, the researchers mainly carried out the mining of opinion at the three levels of document level, statement level and attribute level, and some excellent survey summarized the research status of these levels, but there is no overall overview of these three aspects. Mainly introduces the research status of these three levels of opinion mining;firstly, introduces the definition of each level of opinion mining;then, introduces the methods used in opinion mining;finally, compares the advantages and disadvantages of various methods, also summarizes the use of more emotional dictionaries.
作者
赵泽青
ZHAO Ze-qing(College of Computer Science, Sichuan University, Chengdu 610065)
出处
《现代计算机》
2019年第7期49-53,共5页
Modern Computer
关键词
观点挖掘
情感倾向性分析
挖掘层次
Perspective Mining
Emotional Orientation Analysis
Mining Level
