A Diversity-Accuracy Measure for Homogenous Ensemble Selection

A Diversity-Accuracy Measure for Homogenous Ensemble Selection

Several selection methods in the literature are essentially based on an evaluation function that determines whether a model M contributes positively to boost the performances of the whole ensemble. In this paper, we propose a method called DIversity and ACcuracy for Ensemble Selection (DIACES) using...

Saved in:
Journal Title: International Journal of Interactive Multimedia and Artificial Intelligence
First author: S. Taleb Zouggar
Other Authors: A. Adla
Palabras clave:
Language: Undetermined
Get full text: https://www.ijimai.org/journal/sites/default/files/files/2018/06/ijimai_5_5_8_pdf_37634.pdf
https://www.ijimai.org/journal/node/2460
Resource type: Journal Article
Source: International Journal of Interactive Multimedia and Artificial Intelligence; Vol 5, No 5 (Year 2019).
DOI:
Publisher: Universidad Internacional de La Rioja
Usage rights: Reconocimiento (by)
Subjects: Physical/Engineering Sciences --> Computer Science, Artificial Intelligence
Abstract: Several selection methods in the literature are essentially based on an evaluation function that determines whether a model M contributes positively to boost the performances of the whole ensemble. In this paper, we propose a method called DIversity and ACcuracy for Ensemble Selection (DIACES) using an evaluation function based on both diversity and accuracy. The method is applied on homogenous ensembles composed of C4.5 decision trees and based on a hill climbing strategy. This allows selecting ensembles with the best compromise between maximum diversity and minimum error rate. Comparative studies show that in most cases the proposed method generates reduced size ensembles with better performances than usual ensemble simplification methods.