论文信息 - Building Folk UMLS: An Approach to Finding Meaning of Folk Terms in Medical Domain

Building Folk UMLS: An Approach to Finding Meaning of Folk Terms in Medical Domain

As a medical domain knowledge base, the Unified Medical Language System (UMLS) focuses on formal and professional medical terms; online health forums contain user-generated “folk terms”, which can be used to complement the UMLS vocabulary. In this paper, we propose an approach to detecting folk terms from online discussions and matching their meanings to UMLS concepts. This approach makes connections between expert-built ontology and user-generated taxonomy (folksonomy) based on term distance matching using Google distance measurement. By finding meanings of user-generated folk terms, we will build what we call “folk UMLS” ontology as enrichment to the formal UMLS ontology.

[1] Alan R. Aronson,et al. Effective mapping of biomedical text to the UMLS Metathesaurus: the MetaMap program , 2001, AMIA.

[2] C. Friedman,et al. Using BLAST for identifying gene and protein names in journal articles. , 2000, Gene.

[3] Frank van Harmelen,et al. Using Google distance to weight approximate ontology matches , 2007, WWW '07.

[4] Craig A. Morioka,et al. IndexFinder: A Method of Extracting Key Concepts from Clinical Texts for Indexing , 2003, AMIA.

[5] Steffen Staab,et al. Mining Ontologies from Text , 2000, EKAW.

[6] Michael Krauthammer,et al. Term identification in the biomedical literature , 2004, J. Biomed. Informatics.

[7] Peter Willett,et al. Protein Structures and Information Extraction from Biological Texts: The PASTA System , 2003, Bioinform..

[8] Paul M. B. Vitányi,et al. The Google Similarity Distance , 2004, IEEE Transactions on Knowledge and Data Engineering.

[9] William R. Hersh,et al. A survey of current work in biomedical text mining , 2005, Briefings Bioinform..