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

Title:	Concept and relation extraction techniques in ontology learning for hadith text
Authors:	Mahmood Mohammed Ahmed (P72240)
Supervisor:	Saidah Saad, Dr.
Keywords:	Ontology learning Hadith text Relation extraction English Hadith Dissertations, Academic -- Malaysia
Issue Date:	20-Jan-2016
Description:	Ontology learning such as extraction and acquisition methods is the automatic or semi-automatic creation of ontologies, including extracting the domains terms or candidate concepts and the relationships between those concepts from a corpus of natural language text, and encoding them with an ontology language for easy retrieval. As building ontologies manually is extremely labour-intensive and time consuming, there is great motivation to automate the process. In recent years, the field of ontology learning from text has attracted a lot of attention. Ontology learning from text is a challenging interdisciplinary task that consists of natural language processing, knowledge extraction and ontology construction. Up to now, there are very few researches conducted on ontology learning from Islamic domain. This means that there is currently very little automated support for using knowledge from Islamic literature in semantically-enabled systems. This thesis presents an analysis of techniques for the extraction of ontology from the English Hadith translation text (Muslim book of Hadith). This thesis propose a new hybrid techniques in design and develop a prototype for ontology learning system to demonstrates the feasibility of using some existing ontology learning methods and some improvement in NLP pattern for English Hadith translation text. First, the extraction of concepts and instances is based on the linguistic and statistical approaches, including the part of speech tagging, name entity extraction, shallow parsing , C-Value and NC-value methods. Secondly, this work uses a pattern-based method based on parts-of-speech and shallow parsing patterns for the extraction of taxonomy and non-taxonomy relations. This work adopts several states of art lexico-syntactic patterns for extracting taxonomic relations between extracted concepts. The Hadith gold standard has been constructed from 100 Hadith text of Muslim book for the analysis and evaluation. The results show that C-Value is a good method for automatic extraction of Islamic concept. The results show that the used methods are suitable and effective for both concept extraction and relation extraction. The best results obtained for the extraction of concepts, taxonomy and non-taxonomy relations were 89.94, 83.93 and 89.94 of F1, respectively.,Certification of Master's/Doctoral Thesis" is not available
Pages:	78
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.