Abstract.
As the attribute spaces available to geography expand, new challenges are posed in comprehending and analysing data. This article introduces two complementary approaches to analysis that show promise in addressing data with high attribute dimensionality: inductive learning and visualisation. Whilst neither of these techniques are yet as robust or generally available as many accepted parametric techniques, they are nevertheless able to provide insight, and in the case of inductively-based classifiers and approximation methods, have been shown to outperform traditional approaches in some geographic settings. Some problems with parametric inferential statistics are briefly mentioned, followed by descriptions of inductive and visual analysis methods, and some of the important research that remains to be done before they can take a more prominent role.
Similar content being viewed by others
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Gahegan, M. The case for inductive and visual techniques in the analysis of spatial data. J Geograph Syst 2, 77–83 (2000). https://doi.org/10.1007/s101090050033
Issue Date:
DOI: https://doi.org/10.1007/s101090050033