A survey on Chinese word segmentation technology

This talk will give a comprehensive introduction on Chinese word segmentation (CWS) technologies. The problem and difficulty of CWS will be introduced firstly. Then various CWS methods will be given, which include dictionary-based CWS, generative CWS models, discriminative CWS models, and unsupervised CWS methods. Applications of CWS in information retrieval and machine translation will also be discussed.