Polarity Classification of Short Product Reviews via Multiple Cluster-based SVM Classifiers

While substantial studies have been achieved on sentiment analysis to date, it is still challenging to explore enough contextual information or specific cues for polarity classification of short text like online product reviews. In this work we explore review clustering and opinion paraphrasing to build multiple cluster-based classifiers for polarity classification of Chinese product reviews under the framework of support vector machines. We apply our approach to two corpora of product reviews in car and mobilephone domains. Our experimental results demonstrate that opinion clustering and paraphrasing are of great value to polarity classification.