Induction of diagnostic support rules through data mapping - on the example of the Hepar system
Leon Bobrowski, Hanna Wasyluk · 2007
Induction of similarity measures through data transformations is considered in the paper. Particular attention is paid to the separable linear transformation of multidimensional data on a plane and the data visualization on diagnostic maps. Diagnostic support in the Hepar computer system is based on diagnostic maps. Separable linear transformations in the He par system are based on minimization of the convex and piecewise linear (CPL) criterion functions. The CPL criterion functions give a possibility for a flexible and efficient design of the visualizing transformations which well separates the disease on the diagnostic maps. Key w 0 r d s: decision support, similarity measures, linear transformations, data visu alization