The importance of the lexicon in tagging biological text Export

@article{citeulike:848568, abstract = {A part-of-speech tagger is a fundamental and indispensable tool in computational linguistics, typically employed at the critical early stages of processing. Although taggers are widely available that achieve high accuracy in very general domains, these do not perform nearly as well when applied to novel specialized domains, and this is especially true with biological text. We present a stochastic tagger that achieves over 97.44\% accuracy on MEDLINE abstracts. A primary component of the tagger is its lexicon which enumerates the permitted parts-of-speech for the 10 000 words most frequently occurring in MEDLINE. We present evidence for the conclusion that the lexicon is as vital to tagger accuracy as a training corpus, and more important than previously thought.}, author = {Smith, Lawrence H. and Rindflesch, Thomas C. and Wilbur, John W.}, citeulike-article-id = {848568}, citeulike-linkout-0 = {http://133.11.199.19:8080/-_-http://journals.cambridge.org/action/search#}, citeulike-linkout-1 = {http://dx.doi.org/doi:10.1017/S1351324905003967}, doi = {doi:10.1017/S1351324905003967}, journal = {Natural Language Engineering}, keywords = {nlp, tagging-or-annotation}, posted-at = {2006-09-18 12:00:42}, priority = {2}, publisher = {Cambridge University Press}, title = {The importance of the lexicon in tagging biological text}, url = {http://dx.doi.org/doi:10.1017/S1351324905003967}, year = {2005} }

TY - JOUR ID - citeulike:848568 L3 - citeulike-article-id:848568 N2 - A part-of-speech tagger is a fundamental and indispensable tool in computational linguistics, typically employed at the critical early stages of processing. Although taggers are widely available that achieve high accuracy in very general domains, these do not perform nearly as well when applied to novel specialized domains, and this is especially true with biological text. We present a stochastic tagger that achieves over 97.44% accuracy on MEDLINE abstracts. A primary component of the tagger is its lexicon which enumerates the permitted parts-of-speech for the 10 000 words most frequently occurring in MEDLINE. We present evidence for the conclusion that the lexicon is as vital to tagger accuracy as a training corpus, and more important than previously thought. JF - Natural Language Engineering TI - The importance of the lexicon in tagging biological text PB - Cambridge University Press KW - nlp KW - tagging-or-annotation AU - Smith, Lawrence AU - Rindflesch, Thomas AU - Wilbur, John PY - 2005/// UR - http://dx.doi.org/doi:10.1017/S1351324905003967 ER -