一种语句级细粒度情感倾向性分析算法研究  

RESEARCH ON AN ALGORITHM OF STATEMENT-LEVEL FINE-GRAINED EMOTIONAL TENDENCY ANALYSIS

作　　者：黄高峰[1] 周学广[1] 

机构地区：[1]海军工程大学信息安全系,湖北武汉430033

出　　处：《计算机应用与软件》2015年第4期239-242,共4页Computer Applications and Software

基　　金：国家自然科学基金项目(61100042);湖北省自然科学基金项目(2013CFB441)

摘　　要：针对中文语句的细粒度情感倾向分析问题,构建一个细粒度情感词库。对所含词语进行权重值的计算,并结合中文语句的句式、情感词修饰方法等特点,提出一种基于词库权重、句式权重、修饰权重相结合的语句级细粒度情感倾向性分析算法。实验结果表明,该算法可以准确计算出评论语句的情感倾向值,为进一步研究语句级细粒度情感倾向分类提供了重要的参考依据。In this paper we construct a fine-grained emotion thesaurus for fine-grained emotional tendency analysis problem in Chinese sentences,and calculate the weights of words in the thesaurus.On this basis,combining the features of Chinese sentences patterns, modification methods with emotional words and so on,we present an algorithm of fine-grained emotional tendency analysis on statement-level, which is based on the combination of weights of the thesaurus,the weights of sentences patterns,and the weights of modification. Experimental results show that the algorithm can accurately calculate the emotional tendency values of commenting sentences,and provides the important reference for further study of the classification of fine-grained emotional tendency on statement-level.

关 键 词：细粒度情感 情绪 情感倾向性分析 语句级 权重计算 

分 类 号：TP391.1[自动化与计算机技术—计算机应用技术]