A Chinese Word Segmentation System Based on Cascade Model

This paper introduces the system of Word Segmentation and analyzes its evaluation results in the Fourth SIGHAN Bakeoff . A novel method has been used in the system, which main idea is: firstly, the main problems of WS have been classified, and then a cascaded model has been used to gradually optimize the system. The core of this WS system is the segmentation of ambiguous words and the internal information extraction of unknown words. The experiments show that the performance is satisfying, with the RIV-measure 96.8% in NCC open test in the SIGHAN bakeoff 2007.