Subsumption Reasoning for QoS-Based Service Matchmaking

Abstract : Service-orientation has revolutionized the way applications are constructed and provisioned. To this end, a proliferation of web services is being increasingly available. To exploit such services, an accurate service discovery process is required with a suitable performance focusing both on functional and quality of service (QoS) aspects. In fact, QoS is the main distinguishing factor for the plethora of functionally-equivalent services available in the internet. Accuracy in service discovery comes via exploiting formal techniques and ontologies in particular. Satisfactory performance levels can be reached via using smart methods that intelligently organise the service advertisement space. In this paper, we propose smart ontology-based QoS-aware service discovery algorithms that exploit ontology subsumption as a means of matching QoS requests and offers. These algorithms exploit a variety of methods to structure the service advertisement space. Based on the empirical evaluation conducted, our proposed algorithms outperform the state-of-the-art in certain circumstances. To this end, ontology-based subsumption is indeed a promising technique to realise QoS-based service matchmaking.