2012
EMSS 2012 Proceeding
Variable interaction networks in medical data
Authors: Stephan Winkler, Michael Affenzeller, Gabriel Kronberger, Michael Kommenda, Stefan Wagner, Witold Jacak, Herbert Stekel
Abstract
In this paper we describe the identification of variable interaction networks in a medical data set. The main goal is to generate mathematical models for standard blood parameters as well as tumor markers using other available parameters in this data set. For each variable we identify those variables that are most relevant for modeling it; relevance of a variable can in this context be defined via the frequency of its occurrence in models identified by evolutionary machine learning methods or via the decrease in modeling quality after removing it from the data set. Several data based modeling approaches implemented in HeuristicLab have been applied for identifying estimators for selected tumor markers and cancer diagnoses Linear regression and support vector machines (optimized using evolutionary algorithms) as well as genetic programming.
I3M Scientific Sponsors
I3M Industrial Sponsors
I3M Media Sponsors
