%0 Journal Article
%T Using co-occurrence network structure to extract synonymous gene and protein names from MEDLINE abstracts
%A AM Cohen
%A WR Hersh
%A C Dubay
%A K Spackman
%J BMC Bioinformatics
%D 2005
%I BioMed Central
%R 10.1186/1471-2105-6-103
%X Performance was measured on a test set consisting of about 50,000 abstracts from one year of MEDLINE. Synonyms retrieved from curated genomics databases were used as a gold standard. The system obtained a maximum F-score of 22.21% (23.18% precision and 21.36% recall), with high efficiency in the use of seed pairs.The method performs comparably with other studied methods, does not rely on sophisticated named-entity recognition, and requires little initial seed knowledge.The volume of published biomedical research, and therefore the underlying biomedical knowledge base, continues to grow. The MEDLINE 2004 database is currently growing at the rate of about 500,000 new citations each year [1]. With such growth, it is challenging to keep up-to-date with all of the new discoveries and theories even within one's own field of research. Methods must be established to aid biomedical researchers in making better use of the existing published research and helping them put new discoveries into practical use [2].Text mining and knowledge extraction are ways to aid biomedical researchers in identifying important connections within information in the biomedical knowledge base. A subset of natural language processing (NLP), text mining and knowledge extraction concentrate on solving a specific problem in a specific domain identified a priori. For example, literature searching may be improved by identifying all of the names and symbols used in the literature to identify a particular gene [3], or potential new treatments for migraine may be determined by looking for pharmacological substances that regulate biological processes associated with migraine [4,5].Similar to acronym and abbreviation extraction, which has been studied by several groups [6-8], the problem of gene and protein name synonymy is one that can be addressed with the aid of text mining. Many genes and proteins have multiple names with several orthographic and lexical variants. Gene names are often not used consistentl
%U http://www.biomedcentral.com/1471-2105/6/103