Onto miner: Bootstrapping ontologies from overlapping domain specific web sites

Hasan Davulcu, Srinivas Vadrevu, Saravanakumar Nagarajan

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Scopus citations

Abstract

In this paper, we present automated techniques for bootstrapping and populating specialized domain ontologies by organizing and mining a set of relevant overlapping Web sites provided by the user. We develop algorithms that detect and utilize HTML regularities in the Web documents to turn them into hierarchical semantic structures encoded as XML. Next, we present tree-mining algorithms that identify key domain concepts and their taxonomical relationships. We also extract semi-structured concept instances annotated with their labels whenever they are available. Experimental evaluation for the News, Travel, and Shopping domains indicates that our algorithms can bootstrap and populate domain specific ontologies with high precision and recall.

Original languageEnglish (US)
Title of host publicationProceedings of the 13th International World Wide Web Conference on Alternate Track, Papers and Posters, WWW Alt. 2004
PublisherAssociation for Computing Machinery, Inc
Pages500-501
Number of pages2
ISBN (Electronic)1581139128, 9781581139129
DOIs
StatePublished - May 19 2004
Event13th International World Wide Web Conference on Alternate Track, Papers and Posters, WWW Alt. 2004 - New York, United States
Duration: May 19 2004May 21 2004

Publication series

NameProceedings of the 13th International World Wide Web Conference on Alternate Track, Papers and Posters, WWW Alt. 2004

Other

Other13th International World Wide Web Conference on Alternate Track, Papers and Posters, WWW Alt. 2004
Country/TerritoryUnited States
CityNew York
Period5/19/045/21/04

Keywords

  • Data mining
  • Ontology
  • Semantic web
  • Web mining

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Software

Fingerprint

Dive into the research topics of 'Onto miner: Bootstrapping ontologies from overlapping domain specific web sites'. Together they form a unique fingerprint.

Cite this