Discovering the Research Issues of Classical Chinese Segmentation via Modern Chinese Segmentation System

Word segmentation is the specific issue for Chinese in the natural language processing(NLP) domain. After several decades, there are some Chinese segmentation systems, such as CKIP and Jeiba, which are available for Chinese NLP applications. However, these systems are designed for the modern Chinese, not for the classical Chinese. In this work, we tried to apply the modern Chinese segmentation systems on the classical Chinese articles to find the research issues for the classical Chinese segmentation system. One classical Chinese articles dataset has been carefully selected, processed, and organized as the experimental benchmark by the experts’ professional knowledge.