A Reputation based Trust Model for Geospatial Web Services

Geospatial semantic web services are typically used to process sensory data and produce results for the end user. There are many web services available, providing data products of varying quality. The reliability of services that process or provide this data is of concern since it influences the quality and usefulness of this data. Trust can be used as a means to filter Geospatial Semantic Web services, based on their descriptions and data they provide. Many definitions of trust are provided in the literature and these definitions usually depend on the context and the purpose for which trust is used. We provide a definition of trust for a Web service in a dynamic, open and distributed environment such as the Geospatial Semantic Web. This paper presents progress in developing an ontological representation of trust for Geospatial Web services. The paper highlights challenges in processing geospatial data and presents concepts of a trust ontology for Geospatial Semantic Web services in this context. The paper proposes a reputation based trust model that is dynamic and that takes the service requester's constraints as inputs to the trust function. The trust estimation function presented in this paper is based on direct evidence and indirect evidence of the quality of data produced by the service.