Title: TaxaMiner: an experimentation framework for automated taxonomy bootstrapping

Authors: Vipul Kashyap, Cartic Ramakrishnan, Christopher Thomas, A. Sheth

Addresses: Clinical Informatics R&D, Partners HealthCare System, 93 Worcester St., Wellesley, MA 02481,USA. ' LSDIS Lab, Department of Computer Science, University of Georgia, 415 GSRC, Athens, GA 30602, USA. ' LSDIS Lab, Department of Computer Science, University of Georgia, 415 GSRC, Athens, GA 30602, USA. ' LSDIS Lab, Department of Computer Science, University of Georgia, 415 GSRC, Athens, GA 30602, USA

Abstract: Construction of domain ontologies on the semantic web is a human and resource intensive process, efforts to reduce which are crucial for the Semantic Web to scale. We present a framework for automated taxonomy construction, that involves: (a) generation of a cluster hierarchy from a document corpus using statistical clustering and NLP techniques; (b) extraction of a topic hierarchy from this cluster hierarchy; and (c) assignment of labels to nodes in the topic hierarchy. Metrics for estimating topic hierarchy quality and parameters of an experimentation framework are identified. MEDLINE was the document corpus and MeSH thesaurus was the gold standard.

Keywords: semantic web; domain ontologies; topic hierarchies; taxonomies; ontology learning; automatic taxonomy generation; statistical clustering; natural language processing; taxonomy quality metrics; label assignment; taxonomy bootstrapping; cluster hierarchy; medical literature.

DOI: 10.1504/IJWGS.2005.008322

International Journal of Web and Grid Services, 2005 Vol.1 No.2, pp.240 - 266

Published online: 02 Dec 2005 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article