Semantics-enabled Metadata Generation, Tracking and Validation in Geospatial Web Service Composition for Mining Distributed Images

摘要

In distributed image mining using Web services, Web services implement individual mining algorithms. These services, together with those services providing data and data preprocessing functions, must be chained together dynamically to solve complex real world problems in mining applications. Service composition, the process of creating the service chain, can be divided into three phases: process modeling, process model instantiation, and workflow execution. Semantic Web technologies have been widely used to enable automation in the process-modeling phase. Instantiation of a process model into a concrete workflow or executable service chain requires detailed specification of the metadata for datasets and service instances. This paper explores metadata tracking in instantiating process models. Semantics-enabled metadata are generated and propagated through a service chain. Furthermore, a novel service chaining method based on metadata validation is introduced. Some use cases demonstrate how our approach can contribute to the generation of an executable service chain.