XRVAT: A Visual Topic Maps-Based Environment for Ontology Construction

Abstract

Ontology construction is important in many semantically working environments, such as knowledge management systems and digital libraries. The efficiency, quality, and comprehensiveness of the construction process highly depend on the employed development framework. In this paper, we present a Topic Maps-driven framework called XRVAT for XTM/RDF ontology construction with an effective XTM-to-RDF translation processing kernel. The framework has two major design features. First, the Topic Maps-driven graphical user interface alleviates the cumbersome construction procedure and provides users with a clear top-down conceptual view. Second, the effective XTM-to-RDF translation kernel allows users to incorporate topic features from Topic Maps into RDF and generate XTM/RDF ontologies. Currently, a prototype has been implemented in Java. This paper presents its design methodology and demonstrates its operations with examples. Besides, quantitative analyses were conducted to demonstrate the improvement of the XRVAT interface upon the previous textbox-based interface and its enhanced interface with the visual authoring functionality.