In natural language processing, sentiment lexicon is an effective method for sentiment analysis of text. Aiming at the shortcomings of existing sentiment lexicon in emotional and semantic expression, a method of constructing domain sentiment lexicon based on word2vec is proposed. Based on the existing Chinese affective dictionaries, this method uses TFIDF to measure the importance of vocabulary and the expression of word vectors in semantics to form the selection criteria and classification basis of seed vocabulary, and then forms a sentiment lexicon by category judgment. The comparative experiments show that this method has played a positive role in improving the accuracy of sentiment words classification and the construction of sentiment lexicon.
Paper
Full text
Construction Method of Sentiment Lexicon Based on Word2vec
Semantic Scholar · Computer Science · 2019
Abstract
In natural language processing, sentiment lexicon is an effective method for sentiment analysis of text. Aiming at the shortcomings of existing sentiment lexicon in emotional and semantic expression, a method of constructing domain sentiment lexicon based on word2vec is proposed. Based on the existing Chinese affective dictionaries, this method uses TFIDF to measure the importance of vocabulary and the expression of word vectors in semantics to form the selection criteria and classification basis of seed vocabulary, and then forms a sentiment lexicon by category judgment. The comparative experiments show that this method has played a positive role in improving the accuracy of sentiment words classification and the construction of sentiment lexicon.