engleski

Descriptive induction through subgroup discovery: A case study in a medical domain

This paper presents an approach to descriptive induction based on a combination of subgroup discovery, statistical characterization of induced subgroups, and their visualization. Subgroup discovery is performed using a novel heuristic confirmation rule induction algorithm. Induced subgroup descriptions are enriched by adding statistically significant properties of detected subgroups. Subgroup visualization can be used to present the distribution of detected subgroups in terms of selected properties. The approach is illustrated by the results obtained for the problem of early detection of patient groups with risk for atherosclerotic coronary heart disease.

subgroup discovery; inductive learning

nije evidentirano