Applying ant colony optimization to configuring stacking ensembles for data mining

YiJun CHEN, Man Leung WONG, Haibing LI

Research output: Journal PublicationsJournal Article (refereed)Researchpeer-review

19 Citations (Scopus)

Abstract

An ensemble is a collective decision-making system which applies a strategy to combine the predictions of learned classifiers to generate its prediction of new instances. Early research has proved that ensemble classifiers in most cases can be more accurate than any single component classifier both empirically and theoretically. Though many ensemble approaches are proposed, it is still not an easy task to find a suitable ensemble configuration for a specific dataset. In some early works, the ensemble is selected manually according to the experience of the specialists. Metaheuristic methods can be alternative solutions to find configurations. Ant Colony Optimization (ACO) is one popular approach among metaheuristics. In this work, we propose a new ensemble construction method which applies ACO to the stacking ensemble construction process to generate domain-specific configurations. A number of experiments are performed to compare the proposed approach with some well-known ensemble methods on 18 benchmark data mining datasets. The approach is also applied to learning ensembles for a real-world cost-sensitive data mining problem. The experiment results show that the new approach can generate better stacking ensembles.
Original languageEnglish
Pages (from-to)2688-2702
Number of pages15
JournalExpert Systems with Applications
Volume41
Issue number6
Early online date20 Nov 2013
DOIs
Publication statusPublished - May 2014

Fingerprint

Ant colony optimization
Data mining
Classifiers
Decision making
Experiments
Costs

Bibliographical note

This research is supported by General Research Fund LU310111 from the Research Grant Council of the Hong Kong Special Administrative Region.

Keywords

  • ACO
  • Data mining
  • Direct marketing
  • Ensemble
  • Metaheuristics
  • Stacking

Cite this

@article{1d182b292b7a488fafb5efaa7c92fd4e,
title = "Applying ant colony optimization to configuring stacking ensembles for data mining",
abstract = "An ensemble is a collective decision-making system which applies a strategy to combine the predictions of learned classifiers to generate its prediction of new instances. Early research has proved that ensemble classifiers in most cases can be more accurate than any single component classifier both empirically and theoretically. Though many ensemble approaches are proposed, it is still not an easy task to find a suitable ensemble configuration for a specific dataset. In some early works, the ensemble is selected manually according to the experience of the specialists. Metaheuristic methods can be alternative solutions to find configurations. Ant Colony Optimization (ACO) is one popular approach among metaheuristics. In this work, we propose a new ensemble construction method which applies ACO to the stacking ensemble construction process to generate domain-specific configurations. A number of experiments are performed to compare the proposed approach with some well-known ensemble methods on 18 benchmark data mining datasets. The approach is also applied to learning ensembles for a real-world cost-sensitive data mining problem. The experiment results show that the new approach can generate better stacking ensembles.",
keywords = "ACO, Data mining, Direct marketing, Ensemble, Metaheuristics, Stacking",
author = "YiJun CHEN and WONG, {Man Leung} and Haibing LI",
note = "This research is supported by General Research Fund LU310111 from the Research Grant Council of the Hong Kong Special Administrative Region.",
year = "2014",
month = "5",
doi = "10.1016/j.eswa.2013.10.063",
language = "English",
volume = "41",
pages = "2688--2702",
journal = "Expert Systems with Applications",
issn = "0957-4174",
publisher = "Elsevier Ltd",
number = "6",

}

Applying ant colony optimization to configuring stacking ensembles for data mining. / CHEN, YiJun; WONG, Man Leung; LI, Haibing.

In: Expert Systems with Applications, Vol. 41, No. 6, 05.2014, p. 2688-2702.

Research output: Journal PublicationsJournal Article (refereed)Researchpeer-review

TY - JOUR

T1 - Applying ant colony optimization to configuring stacking ensembles for data mining

AU - CHEN, YiJun

AU - WONG, Man Leung

AU - LI, Haibing

N1 - This research is supported by General Research Fund LU310111 from the Research Grant Council of the Hong Kong Special Administrative Region.

PY - 2014/5

Y1 - 2014/5

N2 - An ensemble is a collective decision-making system which applies a strategy to combine the predictions of learned classifiers to generate its prediction of new instances. Early research has proved that ensemble classifiers in most cases can be more accurate than any single component classifier both empirically and theoretically. Though many ensemble approaches are proposed, it is still not an easy task to find a suitable ensemble configuration for a specific dataset. In some early works, the ensemble is selected manually according to the experience of the specialists. Metaheuristic methods can be alternative solutions to find configurations. Ant Colony Optimization (ACO) is one popular approach among metaheuristics. In this work, we propose a new ensemble construction method which applies ACO to the stacking ensemble construction process to generate domain-specific configurations. A number of experiments are performed to compare the proposed approach with some well-known ensemble methods on 18 benchmark data mining datasets. The approach is also applied to learning ensembles for a real-world cost-sensitive data mining problem. The experiment results show that the new approach can generate better stacking ensembles.

AB - An ensemble is a collective decision-making system which applies a strategy to combine the predictions of learned classifiers to generate its prediction of new instances. Early research has proved that ensemble classifiers in most cases can be more accurate than any single component classifier both empirically and theoretically. Though many ensemble approaches are proposed, it is still not an easy task to find a suitable ensemble configuration for a specific dataset. In some early works, the ensemble is selected manually according to the experience of the specialists. Metaheuristic methods can be alternative solutions to find configurations. Ant Colony Optimization (ACO) is one popular approach among metaheuristics. In this work, we propose a new ensemble construction method which applies ACO to the stacking ensemble construction process to generate domain-specific configurations. A number of experiments are performed to compare the proposed approach with some well-known ensemble methods on 18 benchmark data mining datasets. The approach is also applied to learning ensembles for a real-world cost-sensitive data mining problem. The experiment results show that the new approach can generate better stacking ensembles.

KW - ACO

KW - Data mining

KW - Direct marketing

KW - Ensemble

KW - Metaheuristics

KW - Stacking

UR - http://commons.ln.edu.hk/sw_master/2452

UR - https://www.scopus.com/inward/record.uri?eid=2-s2.0-84890061675&doi=10.1016%2fj.eswa.2013.10.063&partnerID=40&md5=f15f66b6dd1610a96fce3aa47432b1d8

U2 - 10.1016/j.eswa.2013.10.063

DO - 10.1016/j.eswa.2013.10.063

M3 - Journal Article (refereed)

VL - 41

SP - 2688

EP - 2702

JO - Expert Systems with Applications

JF - Expert Systems with Applications

SN - 0957-4174

IS - 6

ER -