Chinese Sentiment Analysis Using Maximum Entropy
Huey Yee Lee, Hemnaath Renganathan · 2011
This paper presents the use of Maximum Entropy technique for Chinese sentiment analysis. Berger, Vincent and Stephen (1996) prove that Maximum Entropy is a technique that is effective in a number of natural language processing applications. In this paper, Maximum Entropy classification is used to estimating the polarity of given comments of from electronic product. These messages are classified into either positive or negative. Apart from presenting the results obtained via Maximum Entropy technique, we also analyze the feature selection and pre-processing of the comments for training and testing purpose. 1