Sememe description based estimating for semantic orientation of Chinese vocabulary

Semantic orientation estimating for Chinese vocabulary has been widely used in text classification, automatic summarization and text filtering, although it is difficult for estimating for Chinese vocabulary, as the rich semantic meaning of chinese vocabulary. The HowNet is a elaborate semantic repository that is helpful for estimating for Chinese vocabulary. This paper proposes using semantic orientation of sememe of the HowNet to estimating semantic orientation of vocabulary. The experiment proves that the method is effective, and it is superior to the method based on the paradigm vocabulary.