摘要
展示了一种新的基于网络评论语言学结构的情感倾向识别模型,固定情感词元模型(fixed sentiment terms model).该方法利用基于固定情感词元的3种特定搭配模式来构造识别算法,通过基于增量的tf-idf模型的相关用户反馈不断更新特征词元集合.通过与传统的情感识别方法相比较,此方法可以较为明显地提高情感分类的效率和准确率.
A new sentimental polarity recognition model based on linguistic structure of emotion states-fixed sentiment terms model was presented.The proposed method used three types of specific collocation pattern to construct the recognition algorithm based on fixed sentiment terms.These feature term sets were gradually updated by relevant feedbacks from the users which based on incremental tf-idf model.Comparison was between the traditional method and fixed sentimental terms model.All tests showed the proposed method got a higher efficiency and accuracy rate of the emotion classifier.
出处
《郑州大学学报（理学版）》
CAS
北大核心
2011年第1期80-84,共5页
Journal of Zhengzhou University:Natural Science Edition
基金
河南省重点科技攻关项目
编号082102210054
郑州市科技攻关项目
编号0910SGYG23259-3
关键词
语言学结构
固定情感词元
增量的tf-idf模型
情感特征选择
情感分类器
linguistic structure
fixed sentimental terms
incremental tf-idf model
sentimental feature extraction
sentimental classifier
