Sugumaran, Vijay and Meziane, F 2009, An architecture to support web‐based information extraction using domain ontologies , in: 8th workshop on EBusiness (WEB2009), , 15 December 2009, Phoenix, Arizona, USA..

PDF Restricted to Repository staff only Download (270kB)

Abstract

The web is the world’s most valuable information resource. However, a wide gap has emerged between the information available for software applications vis-à-vis human consumption. In response to this problem, new research initiatives have focused on extracting information available on the web with machine-processable semantics. Ontologies play a large role in information extraction, particularly in the context of semantic web, and applications should be able to find appropriate ontologies on the fly. However, existing tools do not adequately support information extraction and ontology selection. This research-in-progress paper presents the architecture for an information extraction system which relies on domain ontologies and lexical resources. We also provide an approach for easy identification of appropriate ontologies for a particular task.

Actions (login required)