Chinese comments sentiment classification based on word2vec and SVMperf

作者：

Highlights：

• We achieve similar features clustering using word2vec.

• A method for sentiment classification based on word2vec and SVMperf is proposed.

• Word2vec can extract deep semantic features between words.

• SVMperf trains faster and predicts more accurate than other SVM packages.

• Our classification result can reach more than 90% accuracy.

摘要

•We achieve similar features clustering using word2vec.•A method for sentiment classification based on word2vec and SVMperf is proposed.•Word2vec can extract deep semantic features between words.•SVMperf trains faster and predicts more accurate than other SVM packages.•Our classification result can reach more than 90% accuracy.

论文关键词：Sentiment classification,Word2vec,SVMperf,Semantic features

论文评审过程：Available online 22 October 2014.

论文官网地址：https://doi.org/10.1016/j.eswa.2014.09.011