Spatial Data Modeller (SDM) is a collection of tools for adding categorical maps with interval, ordinal, or ratio scale maps to produce a predictive map of where something of interest is likely to occur. This new release of SDM works for ArcGIS 9.3 and ArcGIS 9.3.1.
A new Home Page for future SDM developments is being implemented here at the University of Campinas by Prof. Carlos Roberto de Souza Filho. This site has been maintained to contain historic versions of SDM and related documentation. It is now being redesigned to contain among other things step-by-step tutorials, new validation tools, possibly new neural net tools such as a Self Organizing Map (SOM) neural net, and other tools. Check out this site for the future evolution of SDM.
Highlights of the new SDM version:
A new tool, Check Data Area, in the SDM Utility toolset quickly checks the area of evidence rasters and reports the area of the Study Area raster.
Three new SDM Utility toolset tools, Clip Categorical Evidence, Clip Ordered Evidence, and Clip Sites, have been added for use in clipping evidence to the Study Area. There are many ways to do this task. Three models are provided as examples of a way to do this clipping
A new approach to multi-class generalization of ordered data is outlined in the documentation for the Calculate Weights tool. This method often can provide a multi-class reclassification of the evidence that has proper weights. This method was suggested by Rob Robinson of the USGS
UNICAMP spatial modeling research team: