Jianqiang Ma is appointed as ESR at the University of Tübingen, Germany from April, 2011 for 2 years.
Situated in the CLARA subproject Finite-State Parsing Methods, Jianqiang Ma’s research topic is Computational Word Formation Model for Chinese Word Segmentation (CWS) and Syntactic Parsing. It combines linguistically motivated rules for grammar development and annotation, with data-driven statistical learning techniques for grammar refinement and parsing.
He has generalized word-based models for CWS to a phrase-based one (Ma and Gerdemann, 2012) in the mainstream framework of sequence labeling. His research has further shown the limitations of such framework (Ma and Gerdemann, 2013) and has proposed a word-structure parsing model to CWS. In order to create the language resource for training this model, he has developed a semi-automatic word structure annotation method (Ma et al., 2012). As to refine the annotation-extracted grammar for more accurate parsing, he has proposed and implemented a novel approach to grammar refinement (Ma et al., 2014). The model will be applied to Chinese word segmentation and syntactic parsing.
The raw annotation and refined grammar will be made publicly available.
Publications in CLARA:
1. [bib] Jianqiang Ma and Dale Gerdemann. Phrase-based approach for adaptive tokenization. In Proceedings of Proceedings of the 12th Meeting of the Special Interest Group on Computational Morphology and Phonology. June 7, 2012. Montreal, Canada.
2. [bib] Jianqiang Ma, Chunyu Kit and Dale Gerdemann. 2012. Semi-automatic annotation of Chinese word structure. In Proceedings of 2nd CIPS-SIGHAN Joint Conference on Chinese Language Processing (CLP-2012). Dec 20-21, 2012, Tianjin, China.
3. Jianqiang Ma and Dale Gerdemann. To appear. Distributional evidence and beyond: the success and limitations of machine learning in Chinese word segmentation. In Proceedings of the 14th Inter- national Conference on Intelligent Text Processing and Computational Linguistics (poster session). March 24-30, 2013. Samos, Greece.
4. Jianqiang Ma et al. 2014. Learning Accurate Grammar of Chinese Word Formation. Forthcoming.