Combining Convolutional Neural Networks and Word Sentiment Sequence Features for Chinese Text Sentiment Classification

Combining Convolutional Neural Networks and Word Sentiment Sequence Features for Chinese Text Sentiment Classification Zhao Chen1, Ruifeng Xu1, Lin Gui1, Qin Lu2 (1. School of Computer Science and Technology, Harbin Institute of Technology Shenzhen Graduate School, Shenzhen, Guangdong, 518000, China; 2. Depart of Computing, The Hong Kong Polytechnic University, Hong Kong, China) Abstract: Recently, the classification approach based on word embedding and convolutional neural networks achieved good results in sentiment classification task. This approach is mainly based on the contextual features of the word embeddings without the consideration of the polarity of the words. Meanwhile, this approach lacks of the use of manually compiled sentiment lexicon resources. Target to these problems, this paper proposes a novel sentiment classification method which incorporates existing sentiment lexicon and convolution neural networks. In this word, the words in text are abstractly represented by using existing sentiment words. The convolutional neural networks are used to extract sequence features from the abstracted word embeddings. Finally, the sequence features are applied to sentiment classification. The evaluations on Chinese Opinion Analysis Evaluation 2014 dataset show that our proposed approach outperforms the convolutional neural networks model with word embedding features and Naïve Bayes Support Vector Machines.