Chinese sentiment classification based on the sentiment drop point

Abstract

The exploding Web opinion data has the essential need for automatic tools to analyze people's sentiments in many fields. Predicting the polarity of a product review is an important work in applications such as market investigation and trend analysis. In this paper, we focus on analyzing the Chinese sentiment word strengths and the sentiment drop point. We propose a novel algorithm based on the sentiment drop point algorithm to conduct sentiment polarity assignment. It predicts the sentiment polarity by a determinative policy which involves two classifiers simultaneously. The experiments show that our approach is efficient and suited for reviews analysis in different domains. © Springer-Verlag Berlin Heidelberg 2013.