Semantic Web Service Based Geospatial Knowledge Discovery

Large amount of Earth and space science data has been collecting from various sources. Effective and efficient knowledge discovery from these distributed multi-disciplinary and multi-scale data is becoming a big challenge. It requires the relevant data and processing steps be discovered, accessed and integrated as much as possible. The Semantic Web provides a common interoperable framework in which information is given well-defined meaning such that the data and operations can be used for more effective discovery and integration across various applications. This paper introduces a new approach to distributed data mining for geospatial knowledge discovery based on semantic Web services and their automatic and semi-automatic chaining. In this approach, domain concepts are well defined by geospatial ontology as the basic knowledge, data and data. mining processes then are well described by these concepts and served by OGC Web services and semantic Web services. So the whole process of geospatial knowledge discovery can be represented as a service chain in predefined patterns of domain concepts. This approach provides an infrastructure that enables individual data and data mining software not only discoverable and accessible, but also interoperable in order to assemble them automatically or semi-automatically to implement more complicated geospatial knowledge discovery.