Abstract:

This paper proposed a new Chinese events recognition method with a variety of common features plus semantic roles features according to the characteristics of Chinese events and using the feature vector to train a CRFs (Conditional Random Fields, CRFs)classifier. When recognizing the Chinese events, this method achieves the F1 measure of 85.4% which increases by 3.6% compared with no semantic role features on the data supplied by the TempEval2 task of SemEval2010. Experimental results show that the proposed method is benefit to Chinese event recognition.

Key words: semantic role features;feature vector;conditional random fields;Chinese events recognition