Abstract
We present OWLS-MX3, the first adaptive hybrid semantic service matchmaker for OWL-S services. It learns how to best combine logic-based, text similarity and ontological structure matching for hybrid semantic selection of OWL-S services to given queries. OWLS-MX3 performs structural semantic matching to compensate for certain cases of text matching failures which are caused by the observed characteristic of many semantic web ontologies of being mere is-a ontologies. The matchmaker adapts its selection to changes in the semantic service landscape and ontologies by learning the respectively optimal weighted combination of different types of semantic similarities it computes for pairs of service requests and service offers. The comparative performance evaluation based on standard measures for both binary and graded service relevance revealed a rather negative result: The improvement of OWLS-MX3 over its non-adaptive predecessor OWLS-MX2 is slight but not significant. On the other hand, its adaptation feature clearly renders OWLS-MX3, in principle, independent from changes of the set of available OWL-S services or matching filters which would otherwise require a manual (re-)combination by the developer to appropriately reflect these changes.