A restricted domain ontology based Arabic question answering for Islamic fatwa

Mustefa Talal Sheker (P72225)

Please use this identifier to cite or link to this item: https://ptsldigital.ukm.my/jspui/handle/123456789/476353

Full metadata record

DC Field	Value	Language
dc.contributor.advisor	Saidah Saad, Dr.
dc.contributor.author	Mustefa Talal Sheker (P72225)
dc.date.accessioned	2023-10-06T09:16:55Z	-
dc.date.available	2023-10-06T09:16:55Z	-
dc.date.issued	2016-02-18
dc.identifier.other	ukmvital:82216
dc.identifier.uri	https://ptsldigital.ukm.my/jspui/handle/123456789/476353	-
dc.description	Automatic Question Answering (QA) is the process of identifying exact answer in response of question asked by the user. Several QA approaches have been proposed for many languages. One of such languages is Arabic which spoken by nearly 200 million people. Approximately 90% of Arab people are Muslims. Many Muslims need religious consultations toward specific issues. Such consultations have to be acquired from a legitimate source such as Islamic scholars. The response toward these consolations is called a Fatwa, which is considered as an Islamic point of view toward such issues. Therefore, there is a vital demand to present a QA system that has the ability to answer Muslims questions based on a collection of Fatwas. There are multiple research efforts have been implemented to provide QA for Islamic domain. However, such efforts have used open-domain knowledge sources such as WordNet, which leads to several limitations in terms of the correctness of the answers. In fact, an open-domain source does not have the ability to identify restricted terms such as the terms occurred in the Islamic domain. In contrast, a restricted-domain knowledge source could serve better due to the definitions and concepts that related to the selected domain are being provided. Hence, this study aims to propose a restricted domain ontology-based question answering approach for the domain of Islamic Fatwa (Prayer book). The ontology has been constructed using a collection of Fatwas, which has been collected from Ibn Uthaymeen-Prayer Fatwas. Several pre-processing tasks have been applied in order to eliminate the irrelevant data (e.g. numbers, non-Arabic letters and punctuation). Furthermore, Term Frequency-Inverse Document Frequency (TFIDF) has been used in order to provide the main concepts of the domain for ontology construction. Such main concepts are occurred frequently in the domain of Islamic thus, TF-IDF could be a useful approach to identify these concepts. The proposed ontology-based QA has been evaluated by comparing it with QA approach that uses simple semantic lexicon as a source of knowledge. The proposed method has outperformed the other approach by obtaining an approximately 90% of F-measure. This demonstrates that the use of ontology as a source of knowledge for QA would enrich the process of extracting the answers.,Certification of Master's/Doctoral Thesis" is not available
dc.language.iso	eng
dc.publisher	UKM, Bangi
dc.relation	Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat
dc.rights	UKM
dc.subject	Domain ontology
dc.subject	Arabic question answering
dc.subject	Islamic fatwa
dc.subject	Dissertations, Academic -- Malaysia
dc.title	A restricted domain ontology based Arabic question answering for Islamic fatwa
dc.type	theses
dc.format.pages	86
dc.identifier.barcode	002329(2016)
Appears in Collections:	Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat

Files in This Item:

File	Description	Size	Format
ukmvital_82216+SOURCE1+SOURCE1.0.PDF Restricted Access		358.11 kB	Adobe PDF	View/Open

Show simple item record Recommend this item