1. UKM Learning and Research Repository
2. Master Thesis / Tesis Master
3. Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat

Title:	Question answering system for hadiths domain concepts using CS and LCS similarity techniques
Authors:	Adnan Hasan (P62450)
Supervisor:	Mohd. Juzaiddin Abaziz, Prof. Dr
Keywords:	Question answering system Hadith Islam Universiti Kebangsaan Malaysia -- Dissertations
Issue Date:	11-May-2014
Description:	Question answering systems are important to retrieve focus and accurate answers based on user’s questions. The main aim of question answering system is to give answers to the user’s questions based on the information that contain documents and information about many fields (open domains) or specific field (restricted domain). The main challenge facing the question answering research is the questions analyzing and extracting to generate accurate and correct answers. This research focuses on Islamic domain and the main aim of the research is to retrieve accurate Hadiths about Islam subjects based on user’s inquiries about Islam concepts. The basic techniques to manage and extract user’s questions are unigram and bigram technique used to generate different possible matching words options of user’s questions and words synonym matching technique used to provide accurate similarity matching between user question and Hadith’s documents based on Hadith’s concepts. Thus, a new list of similarities sentences will be produced to maximize the accuracy of similarity matching. After using synonym technique to generate similarities sentences that are compatible with Hadiths domain concepts, the similarity technique will retrieve the matching answers in Hadith’s schema to display it as answers for user’s queries. Through literature review of theoretical and practical related works, the research domain is classified as restricted domain which is Hadith’s domain, and for the purpose of retrieving accurate answers model of Hadith’s queries as the main objective of this study, the experimental data selected will be based on the dataset quality which consists from 40 famous and trustable Hadiths called Nawawie's Hadiths, and these Hadiths contain many general information and details about the Islam subjects. The prototype test showed that the proposed model provides high accurate performance answers as a result of this research.,Certification of Master's/Doctoral Thesis" is not available
Pages:	78
Call Number:	QA76.9.Q4B394 2014 3 tesis
Publisher:	UKM, Bangi
Appears in Collections:	Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.