Unsupervised Sentiment Orientation Analysis on Micro-Blog Based on Dependency Parsing and Hierarchical Dirichlet Processes
Yong Wang Tang, Bo Wang, Hao Hao Tang, Yao Xi · Applied Mechanics and Materials · 2014
Supervised methods of sentiment orientation analysis have got more and more attention for which don’t require labeled corpus and can be applied to different domains. However, traditional methods mostly consider single words which ignore the dependencies between the words of each other. To solve the problem, this paper presents an unsupervised method for sentiment orientation analysis on micro-blog based on dependency parsing and HDP model. Firstly, this method uses dependency parsing to filter the words in the corpus. Secondly, the HDP model is used to mine the implicit topics in the document. Then, a sentiment dictionary is used to calculate the sentiment distributions of the topics. Finally, the sentiment orientation of the whole micro-blog is obtained on the basis of the sentiment distributions of the topics. The experimental results show that the proposed method can effectively identify the sentiment orientation of micro-blog.