Ontology Score of Malay Translated Hadith as BM25 Ranking Indicator for Malay Text Retrieval

Abstract

Information Retrieval (IR) especially search engine is widely used worldwide. Unfortunately, Information retrieval is lacking in the searching performance. With the usage of ontology, its searching performance can be increased. Ontology is the base of knowledge that keep all the knowledge about the domain. Malay ontology, especially in an Islamic base like Hadith Concept Ontologies on the other hand, are very limited. This research aims to develop a scoring technique using the Malay Translated Hadith Ontology Model as a positive BM25 ranking indicator for Malay Text Retrieval. The results were compared with the original scores of the BM25 Model and the Vector Space Model. The proposed scoring technique outperformed the original BM25 score in 15 queries, while the Vector Space model showed better results in six queries. The ontology score has been found to yield more accurate and superior documents compared to non-ontology search engines, particularly in specific domains or concept areas.