An ontology-based approach to support semantic search in hadith : Zakat domain

Other Title(s)

طريقة معتمدة على الأنطولوجي لدعم البحث الدلالي في الحديث الشريف : مجال الزكاة

Dissertant

al-Misri, Ahmad Jamal Nimr

Thesis advisor

al-Agha, Iyad Muhammad

University

Islamic University

Faculty

Faculty of Information Technology

Department

Information Technology

University Country

Palestine (Gaza Strip)

Degree

Master

Degree Date

2017

English Abstract

The zakat is a tax, comprising percentages of personal income of every kind, It is considered the third pillar of Islam ,The Muslim cannot straight his Islam without performing it when it becomes obligatory on him, the importance of zakat also comes from organizing financial affairs for Muslims , the rules and Laws of distributing wealth on the Muslim society, legislations related to zakat are spread in Islam in the form of quranic verses and prophetic Hadiths, and since quranic verses didn’t cover all the details of zakat it become not enough to understand and explain the zakat fiqh, so hadiths was on of the Islamic resource to explain and understand the zakat fiqh in more details, and without hadith many of Islam pillars become hard to practice, in other hand there is a few of the electronic applications that link the Hadiths with the topics of Islam, especially on the subject of zakat.

We conclude that there an importance of establishing an electronic system that facilitates the access to these Hadiths and retrieving it and to clarifies the relationships between Hadith and how it relate to the zakat pillar.

Islamic domain and content is still not well supported and need to be enriched.

The few works on Islamic Semantic Web applications are still not enough also.

In this research, we propose an ontology based approach for supporting semantic search from Hadiths (zakat field) to enrich Islamic electronic content in retrieving information using semantic web.

It is based on the ontology concept in retrieving the related hadiths to the zakat pillar topics.

Our proposed approach consists of six inter-related components : a knowledge base which was built using ontology that contains the hadiths related to a specific Islamic pillars (zakat).

The reasoning engine which lets user to input queries in Arabic natural language and responses through the I/O interface and uses this dynamic information together with the static knowledge stored in the knowledge base (hadiths) to generate the related hadiths as output.

Ontological dictionary contain all the classes and individuals labels that enables rapid matching of the user input to the knowledge base.

Linguistic module which perform text preprocessing on both input text and ontological dictionary.

The matcher module is responsible for the matching process between the user input and the labels in the ontological dictionary.

SPARQL query module which transfer the matched words from the previous component to a SPARQL query to be handled by linking this query to the zakat ontology to retrieve the related hadiths to the input user which represent the final outputs to the user.

We have evaluated the system using several queries formulated by a domain expert.

This evaluation was carried out by the known metrics (precision and recall).

The results of the evaluation process were as follows: 0.81 for the precision and 0.93 for the recall.

These results were compared by the traditional search which demonstrates a high response to the purpose for which the system was designed and to significantly improve the process of retrieving the Hadiths and greater accuracy in the results.

Main Subjects

Information Technology and Computer Science

No. of Pages

90

Table of Contents

• APA
• MLA
• AMA

Language

English

Data Type

Arab Theses

Record ID

BIM-905188