Exploiting Parallel Texts for Word Sense Disambiguation
Speaker: Hwee Tou Ng , National University of Singapore
Date: April 30 2004
Time: 2:30PM to 3:30PM
Location: Stata Center, 32-D507
Host: Professor Leslie Kaelbling
Contact: Teresa Cataldo, 617-452-5005, cataldo@csail.mit.edu
Relevant URL: Abstract:
Word sense disambiguation (WSD) is the task of determining the correct meaning or sense of a word in context. A central problem of WSD is the lack of manually sense-tagged data required for supervised learning. In this talk, I will present our recent research on investigating the effectiveness of an approach to automatically acquire sense-tagged training data from English-Chinese parallel corpora, first proposed by Resnik and Yarowsky. We trained a word sense disambiguation program for English nouns using automatically sense-tagged English texts from aligned parallel corpora. On the nouns of SENSEVAL-2 English lexical sample task, we compared the WSD accuracy training on automatically sense-tagged English texts, with training on manually sense-tagged examples. On a subset of the most frequently occurring English nouns, we compared the WSD accuracy training on automatically sense-tagged English texts with training on manually sense-tagged examples collected by the Open Mind Word Expert project. Our empirical evaluation reveals that such an approach to acquiring sense-tagged training examples from parallel texts gives promising results.
Bio: Dr Hwee Tou Ng is an Associate Professor of Computer Science at the National University of Singapore and a Senior Faculty Member at the NUS Graduate School for Integrative Sciences and Engineering. He received a PhD in Computer Science from the University of Texas at Austin, USA. His research focuses on natural language processing and information retrieval. He has published papers in premier journals and conferences including Computational Linguistics, ACL, SIGIR, and AAAI. He is an editorial board member of Computational Linguistics journal, a conference organizer and program co-chair of CoNLL-2004 conference, and serves on the program committee of AAAI 2004, ACL 2004, SIGIR 2004, and ICML 2004 conferences.
See other events that are part of
See other events happening in April 2004
