Please use this identifier to cite or link to this item:
https://ptsldigital.ukm.my/jspui/handle/123456789/476341| Title: | Ontology population from English translations of Quran based on lexico-syntactic patterns and association rules |
| Authors: | Taher Weaam Mohammed (P74178) |
| Supervisor: | Saidah Saad, Dr. |
| Keywords: | Quran Ontology creation English translations Lexico-syntactic patterns Dissertations, Academic -- Malaysia |
| Issue Date: | 29-Apr-2016 |
| Description: | Understanding the Quran is a grand challenge for society, public education, Muslimworld education, and also a grand challenge in computational world for knowledge representation and reasoning and for knowledge extraction from text. Understanding from the Quran text is a challenging task due to the nature of the Quran text, which has scattered organization of knowledge and its unique feature. This research aims to extract concept and relation as a part of ontology creation which is based on English translation to the meaning of Quranic text. First, this work used two layers of filtering method, which combines linguistic and statistical methods for concept extraction from two English translation to the meaning of Quran text. Second, this work proposed a new hybrid method based on Lexico-Syntactic patterns and confidence values for Taxonomic relation extraction. Third, this work proposed association rules method for non-Taxonomic relation extraction. To evaluate the proposed methods, this work uses 293 Qurans verses that are related to two Surah; Al-Fatiha and Al-Baqara from two popular English translation of Quran which are the English Quranic translation text by Yusuf Ali and Al-Hilali & Khan Dataset. In addition, an auxiliary has been used in this study which is WordNet ontology in terms of acquiring more related relations. The results showed that using the two layers of filtering method prove to be adequate and efficient measures for automatic extraction of Quranic concepts from both English translation of Quran texts with an overall F-measure of 85.3%. In addition, the results obtained indicate that the proposed methods for both Taxonomic and Non Taxonomic Relation extraction are very suitable technique for extracting relation from both English translation of Quran texts and with an overall F-measure of 87.3% and 88.3% respectively. The proposed method has demonstrated a remarkable improvement in terms of extracting concepts and relations from Quranic English translation.,Certification of Master's/Doctoral Thesis" is not avilable |
| Pages: | 136 |
| Publisher: | UKM, Bangi |
| Appears in Collections: | Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| ukmvital_82204+SOURCE1+SOURCE1.0.PDF Restricted Access | 95.38 kB | Adobe PDF |  View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.