一种基于情感句模的文本情感分类方法  

A Sentiment Classification Approach Based on Sentiment Sentence Framework

作　　者：陈涛[1,2] 徐睿峰[1] 吴明芬[2] 刘滨[1] 

机构地区：[1]哈尔滨工业大学深圳研究生院,广东深圳518055 [2]五邑大学计算机学院,广东江门529020

出　　处：《中文信息学报》2013年第5期67-74,共8页Journal of Chinese Information Processing

基　　金：高等院校博士学科点专项基金资助项目(20122302120070);国家自然科学基金资助项目(11271040);深圳市基础研究计划资助项目(JCYJ20120613152557576;深圳市国际合作计划资助项目(GJHZ20120613110641217);广东省自然科学基金资助项目(S2011010003681);模式识别国家重点实验室开放课题基金资助项目

摘　　要：考虑到同类型的情感句往往具有相同或者相似的句法和语义表达模式,该文提出了一种基于情感句模的文本情感自动分类方法。首先,将情感表达相关句模人工分为3大类105个二级分类;然后,设计了一种利用依存特征、句法特征和同义词特征的句模获取方法,从标注情感句中半自动地获取情感句模。最后,通过对输入句进行情感句模分类实现文本情感分类。在NLP&CC2013中文微博情绪分类评测语料及RenCECps博客语料的实验结果显示,该文提出的分类方法准确率显著高于基于词特征支持向量机分类器。Considering that opinionated sentences always have the same or similar syntax and semantic expression frameworks,this paper proposes a sentiment analysis approach based on sentiment sentence framework.Firstly,we divided sentiment sentence frameworks into three categories and 105subcategories.A sentence framework extraction method is designed to semi-automatically extract sentiment sentence frameworks from annotated sentiment sentences using dependency features,syntactic features and synonym features.The polarity of input sentence is determined through the classification of its sentiment sentence frameworks.The evaluations on NLP＆CC 2013micro-blog emotion analysis corpus and RenCECps blog emotion corpus show that our proposed sentiment classification approach achieves better precision performance compared to word-based support vector machine classifiers.

关 键 词：情感句模 情感分类 句法特征 依存特征 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]