Abstract

This paper presents a method for the visualisation of subgroups, detected by a subgroup discovery algorithm. The main advantage and novelty of the method is that the visualized models can be used to illustrate the distributions of detected groups in terms of the percentages of true positive and false positive cases covered by the model. Subgroup visualization is illustrated by graphs obtained for risk groups discovered in the problem of early detection of patient groups with risk for atherosclerotic coronary heart disease.

Citations