Sentiment Analysis Based on Weibo Comments
Juntao Xue, Xin. Ban, Hao Guo, Xinshan Zhu · 2018
With the rapid development of natural language processing, sentiment analysis has become a very hot research direction. Based on Weibo comments, in order to make up for the defects that the sentiment dictionary cannot update the new words in real time, the paper uses the support vector machine with better classification performance to do the sentiment classification. In this paper, Word2vec is used to train texts of language model. The main innovation is combining Word2vec and TF-IDF to achieve the improvement of word vectors. Then the average of all the improved word vectors in the Weibo sentence is used as the input of support vector machine for emotional analysis. Finally, Word2vec is used to construct the LSTM deep learning model as a comparative experiment. The experimental results show that the proposed method has a higher accuracy than the LSTM model based on Word2vec and avoids the disadvantages of LSTM training model on the complexity of the training model and the long training time.