Research on Sentiment Analysis of Multiple Classifiers Based on Word2vec

Sentiment analysis is a vital application in area of Natural Language Processing (NLP), especially in classification which targets the emotional context of the text. Sentiment analysis can be viewed as a way of quantifying qualitative data using some emotional score indicators.Although emotions belong to category of subjective evaluation, there are many methods in quantitative analysis of emotions such as products evaluation and opinion evaluation. Feature extraction is an important part of sentiment analysis. This paper introduces a feature extraction method called Word2vec. We use Principal Component Analysis (PCA) method is to find the most important elements and structures in the data which is the best dimension. In the end, this paper uses a minimum number of dimensional training data to compare the angles and find the optimal classifier. After extracting features, this paper uses different classifiers for sentiment analysis and uses correct rates to compare the effects of classification.