Incorporating Uncertainty Information into the Visual Analysis of Self-Organizing Maps

GeoViz: Linking Geovisualization with Spatial Analysis and Modeling, 10-11 March 2011, Hamburg, Germany

Tonio Fincke

“Data uncertainty is a frequent issue in data analysis. It describes the possible deviation of a true value from a value which is estimated to be true. In most cases, only this representing value is considered for analysis. Information about uncertainty can be useful to rate the quality of analysis results. It might also help to find alternative explanations for certain patterns or value combinations within a data set[5]. For the description of spatial uncertainties, several visualization methods have been developed, often with the objective of giving the user the possibility to get a good impression of the reliability of the data at a certain point or area in space[1].”