A Data Mining Query Language for Knowledge Discovery in a Geographical Information System

Spatial data mining is a process used to discover interesting but not explicitly available, highly usable patterns embedded in both spatial and nonspatial data, which are possibly stored in a spatial database. An important application of spatial data mining methods is the extraction of knowledge from a Geographic Information System (GIS). INGENS (INductive GEographic iNformation System) is a prototype GIS which integrates data mining tools to assist users in their task of topographic map interpretation. The spatial data mining process is aimed at a user who controls the parameters of the process by means of a query written in a mining query language. In this paper, we present SDMOQL (Spatial Data Mining Object Query Language), a spatial data mining query language used in INGENS, whose design is based on the standard OQL (Object Query Language). Currently, SDMOQL supports two data mining tasks: inducing classification rules and discovering association rules. For both tasks the language permits the specification of the task-relevant data, the kind of knowledge to be mined, the background knowledge and the hierarchies, the interestingness measures and the visualization for discovered patterns. Some constraints on the query language are identified by the particular mining task. The syntax of the query language is described and the application to a real repository of maps is briefly reported.