Semantic enrichment for improving system interoperability

The overall goal addressed in this paper is to improve se- mantic interoperability in heterogeneous systems by means of establishing mappings between relevant domain ontolo- gies. The mappings are discovered based on the technique of semantic enrichment through extension analysis, i.e. using instance information of the ontology to enrich the original ontology and further to calculate similarities between con- cepts in two ontologies. Text categorization is used to au- tomatically assign instance to the concepts in the ontology. Information retrieval techniques are used to calculate simi- larity between concepts. Based on the similarities measure, a heuristic method is used to establish mapping assertions for the two ontologies. The method is illustrated using a product catalogue scenario.