An Ontology Based Approach to Enhance Information Retrieval from Al-Shamelah Digital Librar

Abstract

With the huge number of Islamic resources that emerged over hundreds of years, several difficulties were introduced when searching in this huge heritage. This has challenged developers to build computer applications to facilitate information retrieval from Islamic Resources. One of these applications is Al-Shamelah Digital Library (ADL), Al-maktabah Al-Shamelah, which is a huge database containing thousands of books in different disciplines. The search facility offered by ADL is mainly based on keyword matching, and does not provide semantic interpretations of Islamic texts. It also does not handle complex queries or extract implicit relations and meanings from text. Driven by these challenges, this work presents OntoADL, a system that supports semantic search over a section of Al-Shamelah digital library. At the core of OntoADL is our approach that leverages ontology-based annotations to produce highly relevant search results and to offer recommendations of related topics. The design and architecture of OntoADL is discussed, focusing on how ontology-based reasoning can result in intelligent results that meet the user’s interests. The search service in OntoADL was evaluated by being compared with the search service in the conventional ADL. The OntoADL achieved 83% recall and 66% precision while the ADL system achieved 70% recall and 36% precision.