Extraction of gene-disease relations from Medline using domain dictionaries and machine learning.

by: HW Chun, Y Tsuruoka, JD Kim, R Shiba, N Nagata, T Hishiki, J Tsujii

Pac Symp Biocomput (2006), pp. 4-15.

View FullText article

Pubmed, Hubmed

Reviews [Write a review of this article]

There are no reviews of this article

Find related articles from these CiteULike users

fisherp, bundschu, ibuch, generif_paper (group)

Find related articles with these CiteULike tags

candidategenes, diseasem, extraction, gene, gene-disease, knowledgediscovery, machine-learning, relation, textmining, text-mining

Abstract

We describe a system that extracts disease-gene relations from Medline. We constructed a dictionary for disease and gene names from six public databases and extracted relation candidates by dictionary matching. Since dictionary matching produces a large number of false positives, we developed a method of machine learning-based named entity recognition (NER) to filter out false recognitions of disease/gene names. We found that the performance of relation extraction is heavily dependent upon the performance of NER filtering and that the filtering improves the precision of relation extraction by 26.7% at the cost of a small reduction in recall.

@article{citeulike:1052546, abstract = {We describe a system that extracts disease-gene relations from Medline. We constructed a dictionary for disease and gene names from six public databases and extracted relation candidates by dictionary matching. Since dictionary matching produces a large number of false positives, we developed a method of machine learning-based named entity recognition (NER) to filter out false recognitions of disease/gene names. We found that the performance of relation extraction is heavily dependent upon the performance of NER filtering and that the filtering improves the precision of relation extraction by 26.7\% at the cost of a small reduction in recall.}, address = {Tsujii Laboratory, Room 615, 7th Building of Science, University of Tokyo, Hongo 7-3-1, Bunkyo-ku, Tokyo, 113-0033, Japan. chun@is.s.u-tokyo.ac.jp}, author = {Chun, H. W. and Tsuruoka, Y. and Kim, J. D. and Shiba, R. and Nagata, N. and Hishiki, T. and Tsujii, J. }, citeulike-article-id = {1052546}, journal = {Pac Symp Biocomput}, keywords = {candidategenes, knowledgediscovery, textmining}, pages = {4--15}, posted-at = {2008-04-03 11:45:37}, priority = {2}, title = {Extraction of gene-disease relations from Medline using domain dictionaries and machine learning.}, url = {http://view.ncbi.nlm.nih.gov/pubmed/17094223}, year = {2006} }

RIS record

TY - JOUR ID - citeulike:1052546 L3 - citeulike-article-id:1052546 TI - Extraction of gene-disease relations from Medline using domain dictionaries and machine learning. JF - Pac Symp Biocomput SP - 4 EP - 15 AD - Tsujii Laboratory, Room 615, 7th Building of Science, University of Tokyo, Hongo 7-3-1, Bunkyo-ku, Tokyo, 113-0033, Japan. chun@is.s.u-tokyo.ac.jp N2 - We describe a system that extracts disease-gene relations from Medline. We constructed a dictionary for disease and gene names from six public databases and extracted relation candidates by dictionary matching. Since dictionary matching produces a large number of false positives, we developed a method of machine learning-based named entity recognition (NER) to filter out false recognitions of disease/gene names. We found that the performance of relation extraction is heavily dependent upon the performance of NER filtering and that the filtering improves the precision of relation extraction by 26.7% at the cost of a small reduction in recall. KW - candidategenes KW - knowledgediscovery KW - textmining AU - Chun, HW AU - Tsuruoka, Y AU - Kim, JD AU - Shiba, R AU - Nagata, N AU - Hishiki, T AU - Tsujii, J PY - 2006/// UR - http://view.ncbi.nlm.nih.gov/pubmed/17094223 ER -

RIS BibTeX