Creating a Chinese emotion lexicon based on corpus Ren-CECps

Abstract

As a basic emotion resource, an emotion lexicon plays a very important role in the classification and recognition of emotion in text. This paper proposes an automatic approach to create a Chinese emotion lexicon with tag of emotion intensity based on the emotion corpus Ren-CECps 1.0. We present a new algorithm of emotion vector computation by making use of two language dictionaries Tongyici Cilin and HowNet. Comparing with the experimental results on verbs by SVM classification method, our approach is proved feasible and effective.