Compact dual ensembles for active learning

Amit Mandvikar; Huan Liu; Hiroshi Motoda

doi:10.1007/978-3-540-24775-3_37

Compact dual ensembles for active learning

Amit Mandvikar, Huan Liu, Hiroshi Motoda

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

2 Scopus citations

Abstract

Generic ensemble methods can achieve excellent learning performance, but are not good candidates for active learning because of their different design purposes. We investigate how to use diversity of the member classifiers of an ensemble for efficient active learning. We empirically show, using benchmark data sets, that (1) to achieve a good (stable) ensemble, the number of classifiers needed in the ensemble varies for different data sets; (2) feature selection can be applied for classifier selection from ensembles to construct compact ensembles with high performance. Benchmark data sets and a real-world application are used to demonstrate the effectiveness of the proposed approach.

Original language	English (US)
Title of host publication	Advances in Knowledge Discovery and Data Mining - 8th Pacific-Asia Conference, PAKDD 2004, Proceedings
Editors	Honghua Dai, Ramakrishnan Srikant, Chengqi Zhang
Publisher	Springer Verlag
Pages	293-297
Number of pages	5
ISBN (Print)	354022064X, 9783540220640
DOIs	https://doi.org/10.1007/978-3-540-24775-3_37
State	Published - 2004
Event	8th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2004 - Sydney, Australia Duration: May 26 2004 → May 28 2004

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	3056
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	8th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2004
Country/Territory	Australia
City	Sydney
Period	5/26/04 → 5/28/04

ASJC Scopus subject areas

Theoretical Computer Science
General Computer Science

Access to Document

10.1007/978-3-540-24775-3_37

Cite this

Mandvikar, A., Liu, H., & Motoda, H. (2004). Compact dual ensembles for active learning. In H. Dai, R. Srikant, & C. Zhang (Eds.), Advances in Knowledge Discovery and Data Mining - 8th Pacific-Asia Conference, PAKDD 2004, Proceedings (pp. 293-297). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 3056). Springer Verlag. https://doi.org/10.1007/978-3-540-24775-3_37

Compact dual ensembles for active learning. / Mandvikar, Amit; Liu, Huan; Motoda, Hiroshi.
Advances in Knowledge Discovery and Data Mining - 8th Pacific-Asia Conference, PAKDD 2004, Proceedings. ed. / Honghua Dai; Ramakrishnan Srikant; Chengqi Zhang. Springer Verlag, 2004. p. 293-297 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 3056).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Mandvikar, A, Liu, H & Motoda, H 2004, Compact dual ensembles for active learning. in H Dai, R Srikant & C Zhang (eds), Advances in Knowledge Discovery and Data Mining - 8th Pacific-Asia Conference, PAKDD 2004, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 3056, Springer Verlag, pp. 293-297, 8th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2004, Sydney, Australia, 5/26/04. https://doi.org/10.1007/978-3-540-24775-3_37

Mandvikar A, Liu H, Motoda H. Compact dual ensembles for active learning. In Dai H, Srikant R, Zhang C, editors, Advances in Knowledge Discovery and Data Mining - 8th Pacific-Asia Conference, PAKDD 2004, Proceedings. Springer Verlag. 2004. p. 293-297. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-540-24775-3_37

Mandvikar, Amit ; Liu, Huan ; Motoda, Hiroshi. / Compact dual ensembles for active learning. Advances in Knowledge Discovery and Data Mining - 8th Pacific-Asia Conference, PAKDD 2004, Proceedings. editor / Honghua Dai ; Ramakrishnan Srikant ; Chengqi Zhang. Springer Verlag, 2004. pp. 293-297 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{a9df244bfe2a4fe09dde7e1da4a529c3,

title = "Compact dual ensembles for active learning",

abstract = "Generic ensemble methods can achieve excellent learning performance, but are not good candidates for active learning because of their different design purposes. We investigate how to use diversity of the member classifiers of an ensemble for efficient active learning. We empirically show, using benchmark data sets, that (1) to achieve a good (stable) ensemble, the number of classifiers needed in the ensemble varies for different data sets; (2) feature selection can be applied for classifier selection from ensembles to construct compact ensembles with high performance. Benchmark data sets and a real-world application are used to demonstrate the effectiveness of the proposed approach.",

author = "Amit Mandvikar and Huan Liu and Hiroshi Motoda",

note = "Publisher Copyright: {\textcopyright} Springer-Verlag Berlin Heidelberg 2004.; 8th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2004 ; Conference date: 26-05-2004 Through 28-05-2004",

year = "2004",

doi = "10.1007/978-3-540-24775-3_37",

language = "English (US)",

isbn = "354022064X",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "293--297",

editor = "Honghua Dai and Ramakrishnan Srikant and Chengqi Zhang",

booktitle = "Advances in Knowledge Discovery and Data Mining - 8th Pacific-Asia Conference, PAKDD 2004, Proceedings",

}

TY - GEN

T1 - Compact dual ensembles for active learning

AU - Mandvikar, Amit

AU - Liu, Huan

AU - Motoda, Hiroshi

N1 - Publisher Copyright: © Springer-Verlag Berlin Heidelberg 2004.

PY - 2004

Y1 - 2004

N2 - Generic ensemble methods can achieve excellent learning performance, but are not good candidates for active learning because of their different design purposes. We investigate how to use diversity of the member classifiers of an ensemble for efficient active learning. We empirically show, using benchmark data sets, that (1) to achieve a good (stable) ensemble, the number of classifiers needed in the ensemble varies for different data sets; (2) feature selection can be applied for classifier selection from ensembles to construct compact ensembles with high performance. Benchmark data sets and a real-world application are used to demonstrate the effectiveness of the proposed approach.

AB - Generic ensemble methods can achieve excellent learning performance, but are not good candidates for active learning because of their different design purposes. We investigate how to use diversity of the member classifiers of an ensemble for efficient active learning. We empirically show, using benchmark data sets, that (1) to achieve a good (stable) ensemble, the number of classifiers needed in the ensemble varies for different data sets; (2) feature selection can be applied for classifier selection from ensembles to construct compact ensembles with high performance. Benchmark data sets and a real-world application are used to demonstrate the effectiveness of the proposed approach.

UR - http://www.scopus.com/inward/record.url?scp=7444251779&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=7444251779&partnerID=8YFLogxK

U2 - 10.1007/978-3-540-24775-3_37

DO - 10.1007/978-3-540-24775-3_37

M3 - Conference contribution

AN - SCOPUS:7444251779

SN - 354022064X

SN - 9783540220640

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 293

EP - 297

BT - Advances in Knowledge Discovery and Data Mining - 8th Pacific-Asia Conference, PAKDD 2004, Proceedings

A2 - Dai, Honghua

A2 - Srikant, Ramakrishnan

A2 - Zhang, Chengqi

PB - Springer Verlag

T2 - 8th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2004

Y2 - 26 May 2004 through 28 May 2004

ER -

Compact dual ensembles for active learning

Abstract

Publication series

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this