Sentiment classification using genetic algorithm and Conditional Random Fields
Jian Zhu, Hanshi Wang, Jintao Mao · 2010
Sentiment classification has attracted increasing interest from Natural Language Processing. This paper explores the genetic algorithm to extract the best feature collections from the semantic features of emotional collections. Conditional Random Fields (CRFs) is employed to model the emotional tendency of web pages which are divided into different types of comments, such as positive comments, negative comments and objective comments. Experimental results on both the product reviews and the 1998 People's Daily corpus show that the proposed algorithm works reasonable in the real calculation.