Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
10355497 | Journal of Biomedical Informatics | 2012 | 7 Pages |
Abstract
⺠We study extraction of well formed and high quality biomedical phrases from MEDLINE. ⺠Syntactic and statistical features are employed. ⺠Machine learning is applied using sets of known good phrases. ⺠Over 85% of such extracted candidates are humanly judged to be of high quality.
Related Topics
Physical Sciences and Engineering
Computer Science
Computer Science Applications
Authors
Won Kim, Lana Yeganova, Donald C. Comeau, W. John Wilbur,