An Integrated Pruning Criterion for Ensemble Learning Based on Classification Accuracy and Diversity

Bin Fu; Zhihai Wang; Rong Pan; Guandong Xu; Peter Dolog

doi:10.1007/978-3-642-30867-3_5

An Integrated Pruning Criterion for Ensemble Learning Based on Classification Accuracy and Diversity

Bin Fu, Zhihai Wang, Rong Pan, Guandong Xu, Peter Dolog

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

9 Citations (Scopus)

Abstract

Ensemble pruning is an important issue in the field of ensemble learning. Diversity is a key criterion to determine how the pruning process has been done and measure what result has been derived. However, there is few formal definitions of diversity yet. Hence, three important factors that should be further considered while designing a pruning criterion is presented, and then an effective definition of diversity is proposed. The experimental results have validated that the given pruning criterion could single out the subset of classifiers that show better performance in the process of hill-climbing search, compared with other definitions of diversity and other criteria.

Original language	English
Title of host publication	7th International Conference on Knowledge Management in Organizations : Service and Cloud Computing
Number of pages	12
Volume	172
Publisher	Springer Publishing Company
Publication date	2013
Pages	47-58
ISBN (Print)	978-3-642-30866-6
ISBN (Electronic)	978-3-642-30867-3
DOIs	https://doi.org/10.1007/978-3-642-30867-3_5
Publication status	Published - 2013
Event	7th International Conference on Knowledge Management in Organizations - Salamanca, Spain Duration: 11 Jul 2012 → 13 Jul 2012 Conference number: 7th

Conference

Conference	7th International Conference on Knowledge Management in Organizations
Number	7th
Country/Territory	Spain
City	Salamanca
Period	11/07/2012 → 13/07/2012

Series	Advances in Intelligent Systems and Computing
Volume	172
ISSN	1615-3871

Keywords

Ensemble Learning, Classifier, Ensemble Pruning, Diversity

Access to Document

10.1007/978-3-642-30867-3_5

http://link.springer.com/chapter/10.1007%2F978-3-642-30867-3_5

AUB Link

Search for the material in Aalborg University Library's search engine

MEco: MEco - Medical Ecosystem - Personalized Event-Based Surveillance
Dolog, P., Xu, G., Lage, R. G., Bayyapu, K. R. & Pan, R.
FP7 IST
01/01/2010 → 30/06/2012
Project: Research

Cite this

@inproceedings{f384cb866f1649dfaa6a45a7cf4183aa,

title = "An Integrated Pruning Criterion for Ensemble Learning Based on Classification Accuracy and Diversity",

abstract = "Ensemble pruning is an important issue in the field of ensemble learning. Diversity is a key criterion to determine how the pruning process has been done and measure what result has been derived. However, there is few formal definitions of diversity yet. Hence, three important factors that should be further considered while designing a pruning criterion is presented, and then an effective definition of diversity is proposed. The experimental results have validated that the given pruning criterion could single out the subset of classifiers that show better performance in the process of hill-climbing search, compared with other definitions of diversity and other criteria.",

keywords = "Ensemble Learning, Classifier, Ensemble Pruning, Diversity",

author = "Bin Fu and Zhihai Wang and Rong Pan and Guandong Xu and Peter Dolog",

year = "2013",

doi = "10.1007/978-3-642-30867-3_5",

language = "English",

isbn = "978-3-642-30866-6",

volume = "172",

series = "Advances in Intelligent Systems and Computing",

publisher = "Springer Publishing Company",

pages = "47--58",

booktitle = "7th International Conference on Knowledge Management in Organizations",

address = "United States",

note = "7th International Conference on Knowledge Management in Organizations ; Conference date: 11-07-2012 Through 13-07-2012",

}

Fu, B, Wang, Z, Pan, R, Xu, G & Dolog, P 2013, An Integrated Pruning Criterion for Ensemble Learning Based on Classification Accuracy and Diversity. in 7th International Conference on Knowledge Management in Organizations: Service and Cloud Computing. vol. 172, Springer Publishing Company, Advances in Intelligent Systems and Computing, vol. 172, pp. 47-58, 7th International Conference on Knowledge Management in Organizations, Salamanca, Spain, 11/07/2012. https://doi.org/10.1007/978-3-642-30867-3_5

An Integrated Pruning Criterion for Ensemble Learning Based on Classification Accuracy and Diversity. / Fu, Bin; Wang, Zhihai; Pan, Rong et al.
7th International Conference on Knowledge Management in Organizations: Service and Cloud Computing. Vol. 172 Springer Publishing Company, 2013. p. 47-58 (Advances in Intelligent Systems and Computing, Vol. 172).

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

TY - GEN

T1 - An Integrated Pruning Criterion for Ensemble Learning Based on Classification Accuracy and Diversity

AU - Fu, Bin

AU - Wang, Zhihai

AU - Pan, Rong

AU - Xu, Guandong

AU - Dolog, Peter

N1 - Conference code: 7th

PY - 2013

Y1 - 2013

N2 - Ensemble pruning is an important issue in the field of ensemble learning. Diversity is a key criterion to determine how the pruning process has been done and measure what result has been derived. However, there is few formal definitions of diversity yet. Hence, three important factors that should be further considered while designing a pruning criterion is presented, and then an effective definition of diversity is proposed. The experimental results have validated that the given pruning criterion could single out the subset of classifiers that show better performance in the process of hill-climbing search, compared with other definitions of diversity and other criteria.

AB - Ensemble pruning is an important issue in the field of ensemble learning. Diversity is a key criterion to determine how the pruning process has been done and measure what result has been derived. However, there is few formal definitions of diversity yet. Hence, three important factors that should be further considered while designing a pruning criterion is presented, and then an effective definition of diversity is proposed. The experimental results have validated that the given pruning criterion could single out the subset of classifiers that show better performance in the process of hill-climbing search, compared with other definitions of diversity and other criteria.

KW - Ensemble Learning, Classifier, Ensemble Pruning, Diversity

U2 - 10.1007/978-3-642-30867-3_5

DO - 10.1007/978-3-642-30867-3_5

M3 - Article in proceeding

SN - 978-3-642-30866-6

VL - 172

T3 - Advances in Intelligent Systems and Computing

SP - 47

EP - 58

BT - 7th International Conference on Knowledge Management in Organizations

PB - Springer Publishing Company

T2 - 7th International Conference on Knowledge Management in Organizations

Y2 - 11 July 2012 through 13 July 2012

ER -

Fu B, Wang Z, Pan R, Xu G, Dolog P. An Integrated Pruning Criterion for Ensemble Learning Based on Classification Accuracy and Diversity. In 7th International Conference on Knowledge Management in Organizations: Service and Cloud Computing. Vol. 172. Springer Publishing Company. 2013. p. 47-58. (Advances in Intelligent Systems and Computing, Vol. 172). doi: 10.1007/978-3-642-30867-3_5

An Integrated Pruning Criterion for Ensemble Learning Based on Classification Accuracy and Diversity

Abstract

Conference

Keywords

Access to Document

AUB Link

Projects

MEco: MEco - Medical Ecosystem - Personalized Event-Based Surveillance

Cite this