A post-selection algorithm for improving dynamic ensemble selection methods

TL;DR

This paper proposes PS-DES, a post-selection scheme that improves dynamic ensemble selection by choosing the best ensemble per query using multiple metrics, leading to better accuracy.

Contribution

Introduces PS-DES, a novel post-selection method that enhances dynamic ensemble selection by evaluating and selecting the best ensemble per query.

Findings

PS-DES outperforms individual DES techniques in accuracy

Using accuracy as a selection metric improves ensemble performance

The approach is validated through extensive experiments

Abstract

Dynamic Ensemble Selection (DES) is a Multiple Classifier Systems (MCS) approach that aims to select an ensemble for each query sample during the selection phase. Even with the proposal of several DES approaches, no particular DES technique is the best choice for different problems. Thus, we hypothesize that selecting the best DES approach per query instance can lead to better accuracy. To evaluate this idea, we introduce the Post-Selection Dynamic Ensemble Selection (PS-DES) approach, a post-selection scheme that evaluates ensembles selected by several DES techniques using different metrics. Experimental results show that using accuracy as a metric to select the ensembles, PS-DES performs better than individual DES techniques. PS-DES source code is available in a GitHub repository