Semantic Representation Extraction from Unstructured Arabic Text

One of the crucial tasks in the Semantic Web research is extracting information from unstructured text and converting it into semantic form to be machine understandable. This semantic representation is useful for many purposes such as question answering, summarization and information retrieval. This paper provides a system for converting Arabic text into Resource Description Framework (RDF) semantic format. The proposed system includes syntactical parser that used to extract triples (subject-predicate-object) from preprocessed Arabic text. Moreover, name entity recognition is used to extract entities which mapped with DBpedia to get URIs. Finally, the corresponding RDF representation which captures semantics of Arabic text is generated.