Ontology Extraction Approach for Prophetic Narration (Hadith) using Association Rules

Abstract

Ontologies have been investigated in many artificial intelligence studies including Knowledge Engineering, Natural Language Processing (NLP) and Knowledge Representation. Moreover, ontological models play an important role in Semantic Web application development. Ontologies are used to represent knowledge in a way that makes it understandable by machines as well as humans. This is achieved by encapsulating semantic aspects of the concepts of a certain domain within the ontology. This research paper is concerned with the use of association rules to extract the ontology of prophetic narrations (Hadith). Our approach involves investigating the use of association rules to identify frequent itemsets over concepts that are related to Islamic jurisprudence (Fiqh) from the Sahîh Al-Bukhârî documents by computing correspondence relations using the Apriori algorithm. In particular, the semantic structure of the Sahîh Al-Bukhârî as a knowledge source is exploited to extract a specific domain ontology, while the conceptual relations embedded in this knowledge source are modeled based on the notion of association rules. The domain ontology will offer a powerful representation of prophetic traditional knowledge, and the association rules will express any relation between two classes of connected concepts in the Sahîh Al-Bukhârî collection.