Summary of the paper
| Title | Towards a Learning Approach for Abbreviation Detection and Resolution. |
|---|---|
| Authors | Klaar Vanopstal, Bart Desmet and Véronique Hoste |
| Abstract | The explosion of biomedical literature and with it the -uncontrolled- creationof abbreviations presents some special challenges for both human readers andcomputer applications. We developed an annotated corpus of Dutch medical text,and experimented with two approaches to abbreviation detection and resolution.Our corpus is composed of abstracts from two medical journals from the LowCountries in which approximately 65 percent (NTvG) and 48 percent (TvG) of theabbreviations have a corresponding full form in the abstract. Our firstapproach, a pattern-based system, consists of two steps: abbreviation detectionand definition matching. This system has an average F-score of 0.82 for thedetection of both defined and undefined abbreviations and an average F-score of0.77 was obtained for the definitions. For our second approach, an SVM-basedclassifier was used on the preprocessed data sets, leading to an averageF-score of 0.93 for the abbreviations; for the definitions an average F-scoreof 0.82 was obtained. |
| Language | Statistical and machine learning methods |
| Topics | Corpus (creation, annotation, etc.), Text mining, Statistical and machine learning methods |
Full paper  |
Towards a Learning Approach for Abbreviation Detection and Resolution. |
| Bibtex | @InProceedings{VANOPSTAL10.737, author = {Klaar Vanopstal, Bart Desmet and Véronique Hoste}, title = {Towards a Learning Approach for Abbreviation Detection and Resolution.}, booktitle = {Proceedings of the Seventh conference on International Language Resources and Evaluation (LREC'10)}, year = {2010}, month = {may}, date = {19-21}, address = {Valletta, Malta}, editor = {Nicoletta Calzolari (Conference Chair), Khalid Choukri, Bente Maegaard, Joseph Mariani, Jan Odjik, Stelios Piperidis, Mike Rosner, Daniel Tapias}, publisher = {European Language Resources Association (ELRA)}, isbn = {2-9517408-6-7}, language = {english} } |