Chang, Baobao


Chang, Baobao

Associate Professor

Research Interests: Computational linguistics, natural language processing

Office Phone: 86-10-6276 5810


Chang, Baobao is an associate professor in the Department of Computer Science and technology, School of EECS. He obtained his B.Sc. from Shanxi University in 1992, and Ph.D. from Peking University in 1999 respectively. His research interests include Computational Linguistics, Chinese Language Resource Engineering, Chinese Segmentation, Syntactic and Semantic Parsing.

Dr. Chang has published over 60 high quality academic papers in international conferences and Journals, including top-tier conferences such as ACL, EMNLP, COLING, SIGIR and IJCAI. He has served in the Program Committee of various international conferences including ACL, EMNLP, COLING etc. He is co-authors of two text books and teaching courses on Computational Linguistics. As one of key contributors to the Comprehensive Language Knowledge Base (CLKB) project, he was awarded the Second Class National Scientific & Technical Progress Prize of China in 2011 and the First Class Scientific & Technical Progress Prize of Ministry of Education of China in 2007. He is vice director of the key laboratory of Computational Linguistics (Peking University), Ministry of Education, China and member of both CIPSC (Chinese Information Processing Society of China) and CAAI (Chinese Association for Artificial Intelligence). He is also serving on the editorial board of the Journal of Chinese Information Processing.

Dr. Chang has been in charge of or involved in many research projects including NSFC, 973 programs, 863 project, etc. His research achievements are summarized as follows:

1)  Dependency parsing: Syntactic parsing is a fundamental task for language processing which has been investigated for decades. Challenges of dependency parsing include heavy feature engineering, high computational cost of high-order factorization, inability of recovering long distance dependencies etc. Dr. Chang and his students proposed several deep learning models with which the burden of feature engineering can be dramatically reduced and high order factorization can be avoided without loss of parsing accuracy. 

2)  Shallow Chinese Semantic parsing: Shallow semantic parsing aims at identifying arguments of given predicates in a sentence and assigning semantic role labels. Dr. Chang and his students proposed a semantic chunking approach to Chinese semantic role labeling, with which both the performance and efficiency of Chinese SRL can be improved. He and his students also proposed several deep learning based models to Chinese SRL achieving the state-of-art performance. He and his students also proposed SRL models in which heterogeneous resource or English resources could be effectively used to improve performance of Chinese SRL. 

Chinese Language Resource Engineering: Annotated language resources, such as corpora, machine readable dictionary, constitute indispensable parts of any NLP research and development. Dr. Chang has been long focusing on designing and building Chinese oriented language resources, especially Chinese-English parallel corpora. He is one of key contributors to the Comprehensive Language Knowledge Base (CLKB), which has been the most influential database of language knowledge for Chinese and has been widely used by hundreds of academic institutions and IT companies inside and outside China.