Iterative Schema-based Semantic Matching

Schema/ontology matching is a critical problem in many application domains, such as, schema/ontology integration, data warehouses, e-commerce, web services coordination, Semantic Web, semantic query processing, catalog matching, etc. I view Match as an operator that takes two graph-like structures (e.g., database schemas or ontologies) and produces a mapping between the nodes of the two graphs that correspond semantically to each other. In this PhD proposal, I first present a classification of state of the art matching techniques. I then introduce a new schema-based approach to matching called semantic matching as implemented within the S-Match system. At present, the approach allows handling only tree-like structures (e.g., taxonomies or concept hierarchies). Finally, I discuss the main research goals to be achieved within the PhD thesis: (i) extend the semantic matching approach to allow handling graphs; (ii) extend the semantic matching algorithm for computing mappings between graphs; (iii) develop a theory of iterative semantic matching in order to improve quality and efficiency of matching via iterations; (iv) implement the above stated goals within the S-Match system and conduct the experimental evaluation in different initial settings.