Are there cluster of communities with the same dynamic behaviour?

Abstract

Demographic and economic changes lead to the phenomena of growing and shrinking cities. The issue of this article is to find groups (cluster) of communities with the same dynamic characteristics in Germany. Community Data Mining represents a methodological approach that discovers logical, mathematical and partly complex descriptions of urban patterns and regularities inside statistical data. The approach relies on 12,430 communities and refers to data fromwell known and easily accessible institutions. Emergent SOM is presented as an appropriate and powerful method for clustering and classification. The application of U*-Matrix shows that it is of high value, first, to visualize the structure of highdimensional data and second, to detect meaningful classes. Knowledge Discovery is applied to find a description and recognition of a given set of cluster. The structure and the machine generated explanations were validated mindful of the spatial analyst and yielded a spatial abstraction. Such approaches might lead to a benchmark system for regional policy or to other strategic instruments such as semi or fully automated urban monitoring systems. © Springer-Verlag Berlin Heidelberg 2010.