Please use this identifier to cite or link to this item:
Title: Word sense disambiguation : Scaling up, domain adaptation and application to machine translation
Keywords: word-sense-disambiguation, parallel-texts, domain-adaptation, machine-translation
Issue Date: 31-Oct-2008
Citation: CHAN YEE SENG (2008-10-31). Word sense disambiguation : Scaling up, domain adaptation and application to machine translation. ScholarBank@NUS Repository.
Abstract: The process of identifying the correct meaning (sense) of a word in context, is known as word sense disambiguation (WSD). Current WSD systems lack training examples. In our work, we describe an approach of gathering training examples from parallel texts. Using such examples as part of our training data, we developed systems for the SemEval-2007 coarse-grained and fine-grained English all-words tasks, obtaining excellent results for both tasks. An issue that affects WSD accuracy is that instances of a word drawn from different domains have different sense priors. To address this, we estimate the sense priors of words drawn from a new domain using machine learning methods. We also use active learning to perform effective domain adaptation of WSD systems. Finally, we show for the first time that integrating a WSD system achieves a statistically significant improvement on the translation performance of a state-of-the-art statistical MT system.
Appears in Collections:Ph.D Theses (Open)

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
thesis.pdf750.66 kBAdobe PDF



Page view(s)

checked on Aug 17, 2018


checked on Aug 17, 2018

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.