Developing an Ontology of Concepts in the Qur'an

Abstract

In recent years, there is growing interest in IT for Islamic Knowledge. Researchers in religious studies have started to use ontologies to improve knowledge construction and extraction from religious texts such as the Qur’an and Hadith. An ontology can be used to describe a logical domain theory with very expressive, complex, and meaningful information. Recent research has been done in Arabic language ontology and on holy Qur'an ontology but they are still incomplete. Also there are some other issues including the process used to extract and construct an ontology that needs extra work.  This paper describes our actual and ongoing work in developing an ontology. Our approach is to investigate the applicability of ontology methods of formal Knowledge Representation from Artificial Intelligence and Text Analytics research to capture and represent abstract concepts in the Qur'an. We will implement three ontology methods.  The first one is to elicit or extract the abstract concepts from experts in the domain. Another approach is to semi-automatically extract concepts from text sources from the domain. The last approach is to find existing partial ontologies for the domain, and try to unify and re-use them. Finally to evaluate our general Qur'an ontology, we will investigate practical use of the ontology in a semantic search application. We have implemented the third approach, merging of existing ontologies, The experimental verification result reveals that our proposed merging methods work well, as checked by human expert. A similarity measure was applied for ontology merging, and we report a high accuracy and recall through experiments.