On Boosting the Accuracy of Medical Image Diagnosis through Cooperation between Classifiers

Abstract

Each classifier has special inner workings that conduct to a specific separation between the samples that belong to various classes. Although two classifiers might lead to a similar prediction accuracy, the items from the validation set are not identically classified even if they are trained on the same set of samples. The current research is focused on learning how to use the separations for the validation set as achieved by several classifiers in order to reach a test prediction accuracy that is better than the ones of the individual classification.