Dekang Lin
	Professor Office:   Ath 357 Phone:   1 780 492 9920 Email:   WWW: http://www.cs.ualberta.ca/~lindek

B.Sc. (Computer Science), Tsinghua University, 1985
Ph.D. (Computing Science), University of Alberta, 1992

Natural Language Processing, Machine Learning, Internet Agents

I am interested Natural Language Processing (NLP) and Machine Learning, especially in parsing, statistical language learning, and question answering.

My long term research goal is to develop a theory and a system for NLP that integrate language processing and language learning. The system consists of a processing module and learning module. The initial processing module is endowed with a set of manually coded knowledge. It is then used to process a large amount of natural language text, generating the parse trees and semantic representations for the sentences, for example. The learning module then extracts knowledge from the processed text. Example kinds of knowledge that can be acquired this way include statistics of words and word combinations, idiomatic usages, word similarity, unknown words, etc. Since the learning module takes as input the processed text, instead of raw text, it can acquire more kinds of knowledge with higher accuracy. The newly extracted knowledge can then be integrated into the processing module automatically and enables it to perform better. Such processing-learning cycles constitute a spiral learning process, in which the processing module gradually feed higher quality structures as input to the learning module and the learning module generate more and more accurate knowedge for the processing module.

I developed a principle-based broad-coverage parser for English. It has been used in many applications, such as information extraction and question answering.

SELECTED PUBLICATIONS:

D. Lin. Automatic Identification of Non-compositional Phrases, In Proceedings of ACL-99, pp.317-324. University of Maryland, College Park, Maryland, 1999.

D. Lin. Automatic Retrieval and Clustering of Similar Words. COLING-ACL98, Montreal, Canada, 1998.

D. Lin. An Information-Theoretic Definition of Similarity. Proceedings of International Conference on Machine Learning, Madison, Wisconsin, 1998.

D. Lin. Extracting Collocations from Text Corpora. First Workshop on Computational Terminology, Montreal, Canada, 1998.

D. Lin. Dependency-based Evaluation of MINIPAR. In Workshop on the Evaluation of Parsing Systems, Granada, Spain, 1998.

D. Lin. A Dependency-based Method for Evaluating Broad-Coverage Parsers. Natural Language Engineering, 1998.

D. Lin. Using Syntactic Dependency as Local Context to Resolve Word Sense Ambiguity In Proceedings of ACL-97, Madrid, Spain, 1997.

Publications co-authored with University of Alberta Computing Science students 1997 - present

Patrick Pantel and Dekang Lin. Discovering Word Senses from Text. In Proceedings of ACM SIGKDD Conference on Knowledge Discovery and Data Mining 2002. pp. 613-619. Edmonton, Canada, 2002.

Dekang Lin and Patrick Pantel. Concept Discovery from Text. In Proceedings of Conference on Computational Linguistics 2002. pp. 577-583. Taipei, Taiwan, 2002.

Patrick Pantel and Dekang Lin. Document Clustering with Committees. In Proceedings of ACM SIGIR Conference on Research and Development in Information Retrieval 2002. pp. 199-206. Tampere, Finland, 2002.

Dekang Lin and Patrick Pantel. Discovery of Inference Rules for Question Answering. Natural Language Engineering 7(4):343-360, 2001.

D. Lin and P. Pantel. Induction of Semantic Classes from Natural Language Text. In Proceedings of ACM SIGKDD Conference on Knowledge Discovery and Data Mining 2001. pp. 317-322, 2001.

D. Lin and P. Pantel. DIRT - Discovery of Inference Rules from Text. In Proceedings of ACM SIGKDD Conference on Knowledge Discovery and Data Mining 2001. pp. 323-328, 2001.

P. Pantel and D. Lin. A Statistical Corpus-Based Term Extractor. In: Stroulia, E. and Matwin, S. (Eds.) AI 2001, Lecture Notes in Artificial Intelligence. pp. 36-46. Springer-Verlag, 2001.

P. Pantel and D. Lin. An Unsupervised Approach to Prepositional Phrase Attachment using Contextually Similar Words. In Proceedings of Association for Computational Linguistics 2000. pp. 101-108, Hong Kong, 2000.

P. Pantel and D. Lin. Word-for-Word Glossing with Contextually Similar Words, In Proceedings of ANLP-NAACL 2000, pp.78-85, Seattle, Washington, 2000.

FOR MORE INFORMATION: