Ontology-based semantic metadata extraction approach

In this paper we describes our approach for automatic generation of learning objects' semantic metadata. The extraction process is based on the OBIE (Ontology Based Information Extraction) systems' principles. The input of our approach is a set of IEEE LOM metadata elements in conformance with two requirements. First, each data element must describe the educational content of the learning object. Second, it must be one of the data elements frequently filled by the learning objects' authors and required by most of the LOM application profiles. Concerning the outputs, each one is a couple of a domain concept (from domain ontology) and a degree of pertinence. Moreover we present the details concerning the integration of our approach to learning objects' repositories by taking the COLORS repository as example. In fact the ultimate goal behind our approach is the improvement of repositories' services by offering semantic metadata.