Knowledge Discovery in Spatial Planning Data: A Concept for Cluster Understanding - Computational Approaches for Urban Environments - page 69

knowledge discovery approach. Furthermore, it is necessary to review the results at
different spatial scales (i.e., nationally, regionally, and locally). However, the step
of hypothesis testing was not part of this chapter. Relevant methods will already be
known to readers.
The data on settlement and open space development employed here was bor-
rowed from the publicly available data sets of the IOER Monitor ( http://www.ioer-
monitor.de/ ) . In future, depending on the research question and the availability
of additional variables, the presented illustrative data set on land use (static
perspective) can be expanded, while changes in land use can be characterized in
a multidimensional approach (dynamic perspective). According to the explanatory
power of the parameter land as a barometer for sustainability (Siedentop et al.
2007 ), the static perspective can generate knowledge for the high-level political
evaluation of land use, whereas the dynamic perspective primarily produces prac-
tical information for operative political action at the level of concrete, individual
cases of planning and project-based decisions. In the years to come, the methods
outlined here will find a wide range of application in diverse fields. For example,
they are suited to the machine-based analysis and evaluation of the impact of
spatial programs and measures for urban development imposed at the political level.
Automated spatial survey systems (automated monitoring, classifiers) could assist
in the periodic automatic or semiautomatic estimation and comparison of spatial
objects (reference areas).
Acknowledgements The authors acknowledge the yearly data provided by the Federal Agency for
Cartography and Geodesy, which was crucial for the development of the land use monitoring. The
Federal Institute for Research on Building, Urban Affairs and Spatial Development (BBSR) makes
several spatial typologies available. The authors would like to thank the colleagues of the Leibniz
Institute of Ecological Urban and Regional Development (IOER) for the indicator computation and
the fruitful cooperation. Further, we cordially appreciate the remarks of the reviewers and editors
for giving constructive and helpful comments to improve the quality of this chapter.
Appendix 1
Nonlinear Transformations of the Variables in the UD Data
OpenSpaceMeshSize W log
BuildingArea W log
SettlementDensity W log
SealedSurface W sqrt
LandConsumption W sqrt
ProtectedAreas W sqrt
HemerobyIndex W identity . notransformation /