Wisdom of crowds: An empirical study of ensemble-based feature selection strategies

Teo Susnjak*, David Kerry, Andre Barczak, Napoleon Reyes, Yaniv Gal

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

4 Citations (Scopus)

Abstract

The accuracy of feature selection methods is affected by both the nature of the underlying datasets and the actual machine learning algorithms they are combined with. The role these factors have in the final accuracy of the classifiers is generally unknown in advance. This paper presents an ensemble-based feature selection approach that addresses this uncertainty and mitigates against the variability in the generalisation of the classifiers. The study conducts extensive experiments with combinations of three feature selection methods on nine datasets, which are trained on eight different types of machine learning algorithms. The results confirm that the ensemble based approaches to feature selection tend to produce classifiers with higher accuracies, are more reliable due to decreased variances and are thus more generalisable.

Original languageEnglish
Title of host publicationAI 2015
Subtitle of host publicationAdvances in Artificial Intelligence - 28th Australasian Joint Conference, Proceedings
EditorsJochen Renz, Bernhard Pfahringer
PublisherSpringer-Verlag London Ltd.
Pages526-538
Number of pages13
ISBN (Print)9783319263496
DOIs
Publication statusPublished - 2015
Externally publishedYes
Event28th Australasian Joint Conference on Artificial Intelligence, AI 2015 - Canberra, Australia
Duration: 30 Nov 20154 Dec 2015

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume9457
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference28th Australasian Joint Conference on Artificial Intelligence, AI 2015
Country/TerritoryAustralia
CityCanberra
Period30/11/154/12/15

Fingerprint

Dive into the research topics of 'Wisdom of crowds: An empirical study of ensemble-based feature selection strategies'. Together they form a unique fingerprint.

Cite this