Daniel A. Keim, Christian Panse, Joern Schneidewind, Mike Sips
University of Konstanz
Department of Computer and Information Science
In many application domains, data is collected and referenced by its geo-spatial location. Spatial data mining, or the discovery of interesting patterns in such databases, is an important capability in the development of database systems. A noteworthy trend is the increasing size of data sets in common use, such as records of business transactions, environmental data and census demographics. These data sets often contain millions of records, or even far more. This situation creates new challenges in coping with scale. In this paper we propose a novel pixel-oriented visual data mining approach for large spatial datasets. It combines a quadtree based distortion of map regions and a local reposition of pixels within these map regions to avoid overlap in the display. Experiments shows that it produces visualizations of large data sets for the discovery of local correlations, and is practical for exploring geography-related statistical information in a variety of applications including population demographics, epidemiology, and marketing.