Semi-supervised Chinese Word Segmentation for CLP2012
Saike He, Nan He, Songxiang Cen, Jun Yong Lu · 2012
Chinese word segmentation (CWS) lays the essential foundation for Mandarin Chinese analysis. However, its performance is always limited by the identification of unknown words, especially for short text such as Microblog. While local context are helpless in handling unknown words, global context do manifest enough contextual information, and could be used to guide CWS process. Based on this motivation, in this paper, we report our attempt toward building an integrated model in semi-supervised manner. Considering the complexity of model, we design a strategy to manipulate global and local contextual information asynchronously. Though the coverage of unknown words by such integrated model is still small, official results from CLP2012 present promising result.