الفهرس 
INTRODUCTION.Only 14 pages are availabe for public view
 |  | from | 93 |  |  |
| | | from | 93 | | |
Abstract
In Computer Science, ontology is a shared and common understanding of some domain that can be communicated across people and application systems or enabling knowledge sharing. It is a specification of a conceptualization. The rise of linguistic ontologies is a consequence of two simultaneous circumstances. Data organizing and description to encourage its usage by users later. Language is the best approach to vehicle data and information. So the requirement for semantic information is vital in all research fields. Web information always depends on the language which is written in; the accessibility of data identified with the language that would be much ideal as per the client would be an expanding need for today. Ontology is similar to a structure, dictionary or glossary, however, with more exceptional detail and structure. There is a robust requirement for Arabic language support since the ontology in English cannot be converted into Arabic. Distinctive languages have contained the particular semantic environment and the cultural context, which has brought on the need to build up the other ontology for various languages. Arabic ontology is an important natural language processing field it helps to enrich the Arabic language resources.
The aim of the framework is to automate the process of ontology generation, generating ontologies from pre-existing XML documents. It is an innovative framework, annotated as NAAO (Novel Automatic Arabic Ontology), which automates the ontology generation process from XML documents. The novelty of NAAO resides in generating the Arabic ontology, in the form of XML graph schema (XSG), from semi-structured data (XML documents associated with graph schema). The definition of this automation process was through four main steps necessary to achieve our goal. These steps conclude the main tasks of the automation process for building ontologies from XML documents. This thesis represents a framework that generates an Arabic Ontology from a semi-structured data (XML documents associated with graph schema), in which, XML schema is created and used in the graph schema development (XSG). The thesis provides two case studies, insectivore’s case and mammal’s case study where the developed Arabic ontology is applied. The results consist of 143 words, 10 concepts, 10 elements and 20 relationships. The generated ontology is evaluated using data-driven evaluation methods. 65% of the source XML documents have been included in the insectivore’s case study.
Finally, the thesis provides the implementation of the framework for generating Arabic ontology containing the animal kingdom automatically. The ontology is divided into two parts to be more representative. The first part is the vertebrate’s ontologies which provide 1576 concepts, 3836 element, and 2689 relations. Moreover, the second is the invertebrate’s ontology which contains 320 concepts, 603 elements, and 783 relations. This result can be refined more than one time to reach satisfying results. The generated Arabic ontology is going to be evaluated using data-driven ontology measures cosine similarity measures and tree path mining. Finally, a comparison of the generated framework and three other system is provided.