Mining geo-referenced data with qualitative spatial reasoning strategies

Abstract

The process of knowledge discovery in databases aims at the discovery of associations within data sets.Data Mining is a central step of this process.It corresponds to the application of algorithms for identifying patterns in data.Mining geo-referenced data sets constitutes a special case that demands a particular approach in the knowledge discovery process.Those data sets include allusion to geographic objects,which location and extension de fine implicit relationships of spatial neighbourhood.The Data Mining algorithms have to take this spatial neighbourhood into account when looking for associations among data.This paper presents an approach for knowledge discovery in geo- referenced data sets in which the use of qualitative spatial reasoning strategies makes possible the discovery of patterns that are easily understood by the users.The graphical representation of the results of the knowledge discovery process also allowed a fast understanding of the results achieved.