Authors: Zhan Li, Marek Reformat
Addresses: Electrical and Computer Engineering Department, University of Alberta, Edmonton, T6G 2V4, Canada. ' Electrical and Computer Engineering Department, University of Alberta, Edmonton, T6G 2V4, Canada
Abstract: From the cognitive point of view, web documents consist of a list of concepts instead of terms (words or phrases). Therefore, identifying concepts will provide better results in retrieving useful information from the web. However, web documents are composed of text containing terms and concepts are included implicitly. In order to |find| concepts in a text, we need their definitions and a method for their recognition. In this paper, we propose a novel adaptive assignment of term importance (AATI) schema. This schema is an ontology-based approach for defining and identifying concepts. It includes definitions of relations between terms and concepts, and an iterative algorithm for determining importance of terms. AATI continuously updates importance of terms with |unknown| web documents, which makes it appropriate for web applications.
Keywords: concept definition; concept identification; web documents; ontology; information retrieval; term importance.
International Journal of Computer Applications in Technology, 2010 Vol.38 No.4, pp.333 - 345
Published online: 07 Aug 2010 *Full-text access for editors Access for subscribers Purchase this article Comment on this article