Evaluating the Effectiveness of Thesaurus and Stemming Methods in Retrieving Malay Translated Al-Quran Documents

Information Technology has enabled information in many forms such as text, image or sound, to be accessed widely using search terms via a computer. Due to this type of popularity and advancement in technology, there is an increase interest in searching Malay text to enable scholars and researchers to access databases on-line. Malay texts are scanned are stored in databases ready to be used for text retrieval systems that employ conflation methods to identify word variants from these databases. This paper evaluates the retrieval effectiveness of conflation methods; namely stemming and thesaurus to search and retrieve relevant Malay translated Al-Quran documents based on user natural query words. The Malay Translated Al-Quran texts are stored in an inverted file structure. The retrieved documents are weighted and ranked using Inverse Document Frequency (idf) function. The retrieval effectiveness (E) is measured using standard recall (R) and precision (P). Experiments performed on the Malay Translated Al-Quran documents show that combined search of stemming and thesaurus improve retrieval effectiveness (E) and recall (R) but decrease its precision (P).