Semantics to Energize the Full Services Spectrum: Ontological Approach to Better Exploit Services at Technical and Business Levels

Amit Sheth, Kunal Verma, Karthik Gomadam LSDIS Lab, Dept of Computer Science, University of Georgia {amit, verma, karthik}@cs.uga.edu Introduction Services are pervasive in today's economic landscape, and services-based architectures are being rapidly adopted as IT infrastructure. The need to take a broader perspective of services to include people and organizational descriptions as opposed to technical interface descriptions has already been recognized as part of an overall vision of Services Science [3][5]. The focus of this article is to present the Semantic Services Science (3S) model, which seeks to demonstrate the essential benefits of semantics to the boarder vision of Services Science, with service descriptions that capture technical, human, organizational and business value aspects. We assert that ontology based semantic modeling and descriptions can be used to energize services across the broad service spectrum. In this article, we demonstrate how 3S approach could be used along three points in this spectrum: 1) semantic descriptions of standard Web services with the help of WSDL-S, semantic policies and agreements, 2) semantic descriptions of light weight Web services using Web 2.0 technologies (e.g. REST, AJAX) and 3) ontology based profiling of people and organizational aspects of the assets associated with the knowledge services. Why use semantics as the basis of the service model? Interoperability has been a key challenge in IT for well over a decade. While Electronic Data Exchange standards and XML has provided the basis for data exchange and syntax level interoperability, IT infrastructure needs semantic interoperability to fully exploit and interoperate with respect to all its data and service resources. The fast emerging Semantic Web shows how the use of formal knowledge representation, typically in the form of ontologies, leads to machine processable descriptions, and how the adoption of ontologies that provide common vocabulary and shared knowledge leads to improved semantic interoperability. There are three primary advantages of creating models that employ semantics -1) they promote reuse and interoperability among independently created and managed services, 2) ontology supported representations based on formal and explicit representation lead to more automation, and 3) explicit modeling of the entities and their relationships between them allows performing deep and insightful analysis. For example, if a particular business component is declared as critical, and the relationships between the business and IT components supporting them are explicitly modeled, simple semantic queries may allow business managers to verify if adequate IT resources have been allocated to that component [6]. This reasoning would have otherwise required a systems analyst with great deal of knowledge of both the business values and IT of the organization. 3S utilizes semantic descriptions to capture relationships between services, the people, the organizational aspects, the business values and allows business managers to reason on them to easily create new services and to efficiently allocate resources to the services. On the technical front, much of the modeling effort on services has so far focused on standard Web services in the context of SOA enabled by Web service Description Language (WSDL), SOAP (Simple Object Access Protocol, an XML-based message exchange format) and UDDI (Universal Description, Discovery, and Integration, a technical specification for implementing registries that allow publication and discovery of Web service). It is however possible to take XML-based descriptions used by these standard Web services (and in principle other syntactic descriptions of services) and annotate them with semantics specified in ontologies or conceptual models to gain the above discussed benefits of a semantic approach. The emerging field of Semantic Web Services (SWS), semantics is exploited to discover services using semantic (rather than syntactic) description, to more effectively integrate, compose or orchestrate services to support workflows or processes. The same approach to semantic interoperability can be accorded to fast growing web-based services using Web 2.0 technologies (e.g. REST, AJAX),