Author(s)

M May & A Savinov

Abstract

Data Mining and Geographic Information Systems (GIS) have existed so far as separate technologies. The overall objective of the SPIN! -project is to develop a web-based spatial data mining system by integrating state of the art (GIS) and data mining functionality in a closely coupled open and extensible system architecture. The general architecture for the SPIN! spatial data mining system is described. 1 Introduction It is estimated that 80% of data are gee-referenced. Yet most data mining systems ignore the spatial dimension. On the other hand, contemporary Geographic Information Systems (GIS) have only very basic analysis functionality. This is why data mining combined with GIS offers great potential benefits for solving the problem of spatial data analysis. Although there exist a few attempts to integrate data mining and GIS [4, 11] this is a largely unexplored research area. This paper describes an open, extensible architecture for spatial data mining. It integrates Geographic Information System for interactive visual data exploration and Data Mining functionality specially adapted for spatial data. The architecture of the system will be described, design choices will be discussed. The system is built on the Enterprise Java Bean Architecture (EJB). EJB is a server-side component architecture based on the Java 2 platform. It cleanly separates the \“business logic” (the analysis tools, in our case) from server issues,

Keywords