计算机应用与软件2016，Vol.33Issue(8)：319-322,328,5.DOI:10.3969/j.issn.1000-386x.2016.08.071
基于情感特征向量空间模型的中文商品评论倾向分类算法
CLASSIFICATION ALGORITHM FOR CHINESE PRODUCT REVIEWS TENDENCY BASED ON SENTIMENT FEATURES VECTOR SPACE MODEL
摘要
Abstract
To classify the Chinese product reviews as positive or negative quickly and efficiently,we propose an algorithm.It builds the domain sentiment lexicon in advance according to the review corpus in regard to the products of different categories,and extracts the sentiment features by matching the reviews text with sentiment lexicon set.Then it builds the sentiment feature vector space model (SF-VSM)to solve the problems of traditional vector space model in higher dimensionality and feature selection error.Afterwards,based on SF-VSM and in combination with the improved multinomial naive Bayes method,it classifies the sentiment tendency of reviews.Experimental results show that the proposed algorithm has higher classification accuracy and classification speed than the naive Bayes algorithms based on primitive vector space model or χ2 feature selection respectively.关键词
中文商品评论/情感倾向/情感词典/情感特征向量空间模型/朴素贝叶斯分类Key words
Chinese product reviews/Sentiment tendency/Sentiment lexicon/Sentiment feature vector space model/Naive Bayes classification分类
信息技术与安全科学引用本文复制引用
董祥和..基于情感特征向量空间模型的中文商品评论倾向分类算法[J].计算机应用与软件,2016,33(8):319-322,328,5.基金项目
天津职业技术师范大学科研发展基金项目（SK12-01）。 （）
