Abstract. The increasing interest in Ontologies for many natural language applications in the recent years has led to the creation of ontologies for different purposes and with different feature systems.Manual ontology building is a time consuming activity that requires a lot of effort. In order to overcome these problems many methods have been developed to generate ontologies. There are various studies conducted on Arabic language in Semantic Web. This paper purposes an approach for the generation of an Arabic ontology from semi-structured data (xml document). This approach takes xml document and generates Arabic Ontology. First the system generates xml schema for the xml document. Using this schema it develops the xml schema graph XSG, translate this graph into Arabic and then start the generation of the ontology and the relations using the graph and the xml document. Finally, the system is going to evaluate the generated Arabic ontology using data driven ontology measures.   generates xml schema for the xml document. Using this
schema it develops the xml schema graph XSG, translate this graph into Arabic and then start the
generation of the ontology and the relations using the graph and the xml document. Finally, the system
is going to evaluate the generated Arabic ontology using data driven ontology measures.