Evaluating UMLS strings for natural language processing Export

@article{11825228, abstract = {The National Library of Medicine's Unified Medical Language System (UMLS) is a rich source of knowledge in the biomedical domain. The UMLS is used for research and development in a range of different applications, including natural language processing (NLP). In this paper we investigate the nature of the strings found in the UMLS Metathesaurus and evaluate them for their usefulness in NLP. We begin by identifying a number of properties that might allow us to predict the likelihood of a given string being found or not found in a corpus. We use a statistical model to test these predictors against our corpus, which is derived from the MEDLINE database. For one set of properties the model correctly predicted 77\% of the strings that do not belong to the corpus, and 85\% of the strings that do belong to the corpus. For another set of properties the model correctly predicted 96\% of the strings that do not belong to the corpus and 29\% of the strings that do belong to the corpus.}, address = {National Library of Medicine, National Institutes of Health, USA.}, author = {Mccray, A. T. and Bodenreider, O. and Malley, J. D. and Browne, A. C.}, citeulike-article-id = {4416061}, citeulike-linkout-0 = {http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citation&list_uids=11825228%20}, comment = {1531-605x Journal Article}, journal = {Proc AMIA Symp}, keywords = {controlled, file-import-09-04-28, language, medical, medline, models, natural, processing, statistical, system, unified, vocabulary}, pages = {448--52}, posted-at = {2009-04-28 16:01:51}, priority = {2}, title = {Evaluating UMLS strings for natural language processing}, url = {http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citation&list_uids=11825228%20}, year = {2001} }

TY - JOUR ID - 11825228 L3 - citeulike-article-id:4416061 N2 - The National Library of Medicine's Unified Medical Language System (UMLS) is a rich source of knowledge in the biomedical domain. The UMLS is used for research and development in a range of different applications, including natural language processing (NLP). In this paper we investigate the nature of the strings found in the UMLS Metathesaurus and evaluate them for their usefulness in NLP. We begin by identifying a number of properties that might allow us to predict the likelihood of a given string being found or not found in a corpus. We use a statistical model to test these predictors against our corpus, which is derived from the MEDLINE database. For one set of properties the model correctly predicted 77% of the strings that do not belong to the corpus, and 85% of the strings that do belong to the corpus. For another set of properties the model correctly predicted 96% of the strings that do not belong to the corpus and 29% of the strings that do belong to the corpus. JF - Proc AMIA Symp AD - National Library of Medicine, National Institutes of Health, USA. EP - 52 TI - Evaluating UMLS strings for natural language processing SP - 448 KW - controlled KW - file-import-09-04-28 KW - language KW - medical KW - medline KW - models KW - natural KW - processing KW - statistical KW - system KW - unified KW - vocabulary N1 - 1531-605x Journal Article AU - Mccray, AT AU - Bodenreider, O AU - Malley, JD AU - Browne, AC PY - 2001/// UR - http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citation&list_uids=11825228%20 ER -