Sentiment analysis method based on context view
Chengyun Liu, Liancheng Xu, Zhihao Wang, Ningning Wang · 2016
In recent years, with the continuous development and transmission of micro-blogs. people do more research on analyzing the emotional tendency of short texts. Ignoring the context information, traditional emotional analysis methods on micro-blog are limited in a tweet itself. To tackle this problem, this paper proposes a method of emotional tendency analysis which based on view of the context. Firstly, we establish emotions dictionary of micro-blogs and then add popular expressions among micro-blogs into the dictionary. Secondly, micro-blogs texts are categorized into parts, namely text section, emotions section, and the context related to a tweet. Finally, we carry out sentiment analysis on these three parts. Experimental results show that, compared with traditional machine learning methods, our method in emotional tendency analysis has significant improvement in accuracy.