摘要
近年来,随着互联网、社交媒体、电子商务以及移动终端的高速发展,越来越多的人将自己的观点、心得、意图和评论分享到互联网上,可获取的带有情感倾向性的文本数据急剧增加,文本情感分析也越来越受到研究者的关注。主要从广义的文本情感分析的研究背景和研究价值入手,对当前国内外研究现状进行分析。首先,介绍广义文本情感分析的概念和特点;然后,从任务的角度将广义文本情感分析分为狭义文本情感分类、特定任务类型的情感分类方法、情感元素抽取、立场分析、情感生成、观点总结与情感摘要生成、意图挖掘、情感原因发现及谣言检测与评价真实性分析9个方面,分别介绍和分析上述9个方面的国内外研究进程和现状,帮助读者从宏观的角度上了解广义文本情感分析的主要任务及其研究方法;最后,总结阐述广义文本情感分析存在的主要问题,并结合这些问题对广义文本情感分析未来的研究方向进行展望。
In recent years,with the rapid development of the Internet,social media,e-commerce and mobile devices,more and more people are willing to share their opinions,experiences,intentions and comments on the Internet.Most of these resources are available online.Text sentiment analysis has received increasing attention from researchers.In this paper,the related works in the area of generalized text sentiment analysis were reviewed based on their research background and value.Firstly,the concept and features of generalized text sentiment analysis were introduced.The 9 tasks of narrow text sentiment classification were reviewed,which are sentiment classification method of specific task type,sentiment information extraction,stance detection,sentiment text generation,opinion summary and sentiment summary generation,intention mining,sentiment cause detection and rumor detection.Finally,the problems on generalized text sentiment analysis were summarized,and suggestion on future research of generalized text sentiment analysis based on these problems was given.
作者
杨开漠
吴明芬
陈涛
YANG Kaimo;WU Mingfen;CHEN Tao(Intelligent Manufacturing Division,Wuyi University,Jiangmen Guangdong 529000,China)
出处
《计算机应用》
CSCD
北大核心
2019年第S02期6-14,共9页
journal of Computer Applications
基金
广东省教育厅重大项目(2014KZDXM055)
广东省科技厅项目(2016A030310003)
江门市基础与理论科学研究类科技计划项目(2018JC01003)
研究生教育创新计划项目(2016SFKC_42)
关键词
广义文本情感分析
狭义文本情感分类
情感元素抽取
立场分析
意图挖掘
generalized text sentiment analysis
narrow text sentiment classification
sentiment information extraction
position analysis
intention mining
