Optimizing Classifier Performance in Word Sense Disambiguation by Redefining Sense Classes
Abstract
Learning word sense classes has been shown to be useful in fine-grained word sense disambiguation. However, the common choice for sense classes, WordNet lexicographer files, are not designed for machine learning based word sense disambiguation. In this work, we explore the use of clustering techniques in an effort to construct sense classes that are more suitable for word sense disambiguation end-task. Our results show that these classes can significantly improve classifier performance over the state of the art results of unrestricted word sense disambiguation. URL: http://www.comp.nus.edu.sg/~upali/pub/ijcai-07.pdf
Cite
Text
Kohomban and Lee. "Optimizing Classifier Performance in Word Sense Disambiguation by Redefining Sense Classes." International Joint Conference on Artificial Intelligence, 2007.Markdown
[Kohomban and Lee. "Optimizing Classifier Performance in Word Sense Disambiguation by Redefining Sense Classes." International Joint Conference on Artificial Intelligence, 2007.](https://mlanthology.org/ijcai/2007/kohomban2007ijcai-optimizing/)BibTeX
@inproceedings{kohomban2007ijcai-optimizing,
title = {{Optimizing Classifier Performance in Word Sense Disambiguation by Redefining Sense Classes}},
author = {Kohomban, Upali Sathyajith and Lee, Wee Sun},
booktitle = {International Joint Conference on Artificial Intelligence},
year = {2007},
pages = {1635-1640},
url = {https://mlanthology.org/ijcai/2007/kohomban2007ijcai-optimizing/}
}