Related Experiment Videos
From syntactic-semantic tagging to knowledge discovery in medical texts
W Ceusters1, P Spyns, G De Moor
1Language and Computing NV, Het Moorhof, Zonnegem, Belgium. werner.ceusters@rug.ac.be
International Journal of Medical Informatics
|December 16, 1998
Summary
The MultiTALE system successfully analyzed 81% of surgical procedure expressions from SNOMED International. Further development requires an interface ontology for greater domain and language independence.
Area of Science:
- Medical Informatics
- Natural Language Processing
- Computational Linguistics
Background:
- The GALEN project aims to extract knowledge from surgical procedure expressions.
- MultiTALE is a syntactic-semantic tagger designed for this purpose.
Purpose of the Study:
- To describe the methodology for upgrading MultiTALE.
- To evaluate its performance in analyzing SNOMED International surgical procedure expressions.
Main Methods:
- Upgrading the MultiTALE tagger.
- Analyzing a random sample of surgical procedure expressions from the SNOMED International procedure axis.
Main Results:
- MultiTALE correctly analyzed 81% of the selected expressions.
- Encountered issues included unusual grammar, insufficient domain knowledge, and differing categorizations between domain and linguistic models.
Conclusions:
- The MultiTALE system can assign meaning to previously unencountered words.
- An interface ontology is necessary to enhance independence from specific languages and domains.