... empirical study of Chinese chunking. We compared the performance of four models, SVMs, CRFs, MBL, and TBL.We also investigated the effects of using differentsizes of training data. In order ... generated different sizes of training sets, including 1%, 2%, 5%, 10%, 20%,50%, and 100% of the total training data. In our experiments, we used all the default pa-rameter settings of the packages. ... 110Num of Sentences 9,878 5,290Num of Words 238,906 165,862Num of Phrases 141,426 101,449Table 2: Information of the CTB4 Corpus3 Chinese Chunking3.1 Models for Chinese Chunking In this...