Building ontological relationships: A new approach

  • Authors:
  • Ahmad Kayed;Nael Hirzallah;Luai A. Al Shalabi;Mohannad Najjar

  • Affiliations:
  • Department of Computer Science, Applied Science University, Amman 11931, Jordan;Department of Computer Science, Applied Science University, Amman 11931, Jordan;Department of Computer Science, Applied Science University, Amman 11931, Jordan;Department of Computer Science, Applied Science University, Amman 11931, Jordan

  • Venue:
  • Journal of the American Society for Information Science and Technology
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

Ontology plays an essential role in recognizing the meaning of the information in Web documents. It has been shown that extracting concepts is easier than building relationships among them. For a defined set of concepts, many existing algorithms produce all possible relationships for that set. This makes the process of refining the relationships almost impossible. A new algorithm is needed to reduce the number of relationships among a defined set of concepts produced by existing algorithms. This article contributes such an algorithm, which enables a domain-knowledge expert to refine the relationships linking a set of concepts. In the research reported here, text-mining tools have been used to extract concepts in the domain of e-commerce laws. A new algorithm has been proposed to reduce the number of extracted relationships. It groups the concepts according to the number of relationships with other concepts and provides formalization. An experiment and software have been built, proving that reducing the number of relationships will reduce the efforts needed from a human expert. © 2008 Wiley Periodicals, Inc.