Microblog Emotional Analysis Based on TF-IWF Weighted Word2vec Model
Hao Tian, Liuai Wu · 2018
In order to solve the problem of ignoring the importance of words and missing semantic relations between words in the emotional analysis of microblog, the TF-IWF weighted Word2vec model was proposed as a feature extraction method, and then use Support Vector Machine (SVM) to obtain more accurate results. Firstly, TF-IWF is used to calculate the word weight. Then, Word2vec is used to calculate the words vector, and the weighted word vector is obtained by combining them. Finally, the data is trained and classified through SVM. Experimental results show that compared with the original TF-IWF classification method and Word2vec classification method, the precision and recall of this method are improved.