Paper: Chinese word segmentation model using bootstrapping

ACL ID W10-4136
Title Chinese word segmentation model using bootstrapping
Venue Joint Conference on Chinese Language Processing
Session Main Conference
Year 2010
Authors

We participate in the CIPS-SIGHAN- 2010 bake-off task of Chinese word segmentation. Unlike the previous bakeoff series, the purpose of the bakeoff 2010 is to test the cross- domain performance of Chinese seg- mentation model. This paper summa- rizes our approach and our bakeoff re- sults. We mainly propose to use χ 2 sta- tistics to increase the OOV recall and use bootstrapping strategy to increase the overall F score. As the results shows, the approach proposed in the paper does help, both of the OOV re- call and the overall F score are im- proved.