Chinese sentiment classification of emotional ontology and word vector based on rules

With the rapid development of e-commerce, the research on the emotional tendency of comment texts has attracted the attention of scholars. In order to make full use of the emotional ontology and semantic information in short texts, a Chinese sentiment classification method combining syntactic rules, emotional ontology and word vector is proposed. Word2vec is firstly used to train word vectors and generate short text vectors in conjunction with syntactic rules; Then, according to the distribution of emotional features, a domain adaptive sentiment dictionary is created, and syntactic rules are combined to obtain short text sentiment values, thereby constructing an emotional model called VWEO(Vector with Emotional Ontology) combining word vectors and emotional values. In the hotel review data set, compared with the existing methods, the proposed method has significantly improved accuracy, recall rate and F1 value.