A sentiment classification method for Chinese document

For the great demand of analyzing and mining user´s opinion, this paper presents a novel sentiment classification method for Chinese document. Firstly, the subjective documents are pre-processed and the sentiment features are extracted; then it calculates the basic polarity of the sentiment features based on a sentiment word dictionary, meanwhile dynamic sentiment word´s polarity is judged and polarity adjustment is carried out according to the context information; finally, the overall polarity of document is determined by all sentiment words in it. Experiment of comparing with BaseLine and SVM shows the effectiveness of the proposed method.