Building ontologies: A state of the art, and an application to Finance domain

Abstract

In this paper, we develop a framework for comparing ontology leaning systems and place a number of the more prominent ones into it. We have selected 11 specific projects for this study, including automatic and semi-automatic domain ontology building from unstructured or semi-structured text document. The comparison framework includes general characteristics, such as the purpose of ontology, its coverage (general or domain specific), and the formalism used. It also includes the design process used in creating ontology and the methods used to evaluate it. We synthesize the characteristics of the studied algorithms as a comparison matrix based on predefined criteria. Finally, we selected three methodologies for ontology construction: HCHIRSIM [1], Sanchez et al. 2004 [2] and Text2onto [3] and apply them in Finance domain ontology constructed from the web. We compare the obtained concepts with those on the “Financial glossary” provided by Yahoo.com.