An automated semantic annotation based-on Wordnet ontology

The construction of metadata which annotates the documents is one of the major tasks for making data understandable to the machine on the Semantic Web. Most of the exiting Semantic Web authoring tools allow user with limited knowledge Semantic Web technologies such as RDF, OWL to markup documents with semantics. These tools help to create the semantic annotations to the documents either during or after composing the documents. The common is these approaches all need human annotator to manually determine the semantic information of the content and then create the metadata which describes this information via tools. This research proposes an approach to bridge the gap between today´s Web and the Semantic Web. The word sense disambiguation technique is utilized to automatically discover the underlying semantic information of the text document, based on the popular open-domain vocabulary ontology - the Wordnet. This semantic information is then annotated to the documents in RDF that conforms to the Semantic Web standards for future reuse. We believe this is one of the feasible ways to bring a large-scale of documents to the Semantic Web.