Rules Extraction and Representation for Geographic Information Systems (Short Abstract)

With the rapid use of the World Wide Web a huge amount of spatial data are available on line obtained from different sources and in different time frames. It is, therefore, important for the users to understand the nature of available data in order to use them appropriately for a specific task. In this paper we investigate knowledge representation and sharing in Geographic Information Systems (GIS). Knowledge extraction and representation is a known problem in Artificial Intelligence (AI). Machine learning techniques have been used in different kinds of data but there is very little work regarding spatial data. There is some work proposed in which tools from AI are integrated in a GIS (Albert, 1988) and other approach represents machine learning algorithm for spatial data classification from a GIS (Stearns, 1995). Most data, however, are not integrated in one GIS but are present in several. In this paper we propose a method based on machine learning to extract rules from spatial databases and to import them in a readable form in a GIS. We use the Web Ontology Language (OWL) as a special language for semantic information and the Resource Description Framework Language (RDF) for rules representation and for making them available via the World Wide Web. A