A New Technique for Combining Multiple Classifiers using The Dempster-Shafer Theory of Evidence

TL;DR

This paper introduces a novel classifier combination method based on Dempster-Shafer theory, which adapts to training data to minimize mean square error and outperforms existing methods across multiple classification tasks.

Contribution

The paper proposes a new adaptive classifier combination technique using Dempster-Shafer theory that improves performance over existing methods.

Findings

Outperforms most existing classifier combination methods

Effective on three different classification problems

Minimizes mean square error through adaptation

Abstract

This paper presents a new classifier combination technique based on the Dempster-Shafer theory of evidence. The Dempster-Shafer theory of evidence is a powerful method for combining measures of evidence from different classifiers. However, since each of the available methods that estimates the evidence of classifiers has its own limitations, we propose here a new implementation which adapts to training data so that the overall mean square error is minimized. The proposed technique is shown to outperform most available classifier combination methods when tested on three different classification problems.