Service Network: An Infrastructure Based On Semantics And Social Relations For Web Services Computing

With the development of Internet and widespread of relevant standards of Web Services, Service-Oriented Architecture(SOA) has become the best option when building Enterprise Application Integration(EAI) or Business to Business(B2B) applications on the Web. However, great difficulty exists in information sharing and process mediation when processing service discovery, service combination and interaction in Services-Oriented Computing (SOC).This thesis aims at designing an infrastructure providing better supports for service computing. We use the method of organizing valid Web Services, which own abundant semantic information and socialized relations within interoperation, into services ecosystem, named Service Network (SN). Meanwhile, we conduct further research on the organization and techniques of SN, focusing on semantics and social relations in SN and also on gaining its function in service computing. Some specific contributions of our work are listed as follows:1) Propose the concepts of Service Network. Combining the Semantic Web technology, such as Ontology, and social relations such as reliance, interaction, attribution and ability among services, we present a novel Web services registry, named Service Network to bridge the gap between service consumer and provider. Thus, we present its concepts, components and design the service infrastructure, based on semantics and interactive relations, as a dynamic service ecosystem and social network.2) Analyze, describe and define Service Relations. We take further analysis on the probable relations among three different granularities of Web services, including parameter level, operation level and Web-service level. And then we define several service relations, including mem, sim, equ, sub, plu, seq, pre, com, ref, inv, etc. Meanwhile, we provide detailed implementation of these service relations based on OWL DL.3) Extend the Web services computing model. We add the role of services aggregator, with the responsibility of distributed Web Service interaction and process medium; set Service Network as standardized knowledge base, providing intellectual support for developing and reusing Web services, and also providing entrance for SOA governance. 4) Design and implement the Service Network prototype. We use Semantic Web techniques including Ontology to design a Services Ontology model, which can describe individual attributes and Service Relations within Web services. Based on this, we accomplish the implementation of Service Network prototype.5) Analyze the social relations bewteen Web services. Making full use of classic method of data mining and text analyzing, we analyze the relations of published Web services and network characteristics of various relations among valid services, consequently, validate and guide the design and implementation of Service Network prototype.In a word, we focus on organizing available Web services into service ecosystem with abundant semantic information, business context and interaction relations. We design Service Network with semantics and relations, and introduce semantic reasoning and relational calculus into the combination of the concepts of social networking and semantic Web techniques, and finally develop Service Network as the new basic point to support Services-Oriented Computing.