Related Experiment Videos
Developing a test collection for biomedical word sense disambiguation
M Weeber1, J G Mork, A R Aronson
1Lister Hill National Center for Biomedical Communications, National Library of Medicine, Natioanl Institutes of Health, Bethesda, MD 20894, USA. weeber@nlm.nih.gov
Proceedings. AMIA Symposium
|February 5, 2002
Summary
Resolving word sense ambiguity in biomedical text is crucial for Natural Language Processing (NLP) systems. This study introduces a new Word Sense Disambiguation (WSD) test collection to evaluate NLP techniques in this domain.
Area of Science:
- Computational linguistics
- Biomedical informatics
Background:
- Word sense ambiguity presents a significant challenge for Natural Language Processing (NLP) systems.
- Accurate word sense disambiguation (WSD) is essential for improving NLP system performance, particularly in specialized domains.
Purpose of the Study:
- To develop a specialized test collection for evaluating Word Sense Disambiguation (WSD) techniques.
- To facilitate the advancement of NLP systems within the biomedical language domain.
Main Methods:
- Creation of a novel Word Sense Disambiguation (WSD) test collection.
- The collection includes 5,000 unambiguous instances derived from 50 ambiguous strings within the UMLS Metathesaurus.
Main Results:
- A comprehensive test collection for WSD in the biomedical domain has been established.
- This resource enables rigorous testing and comparison of WSD algorithms.
Conclusions:
- The developed WSD test collection is a valuable resource for advancing NLP in biomedical text.
- Improved WSD techniques will enhance the performance of NLP systems handling complex biomedical terminology.