Least squares support vector machines with fast leave-one-out AUC optimization on imbalanced prostate cancer data

Wang, G; Teoh, JYC; Lu, J; Choi, KS

Least squares support vector machines with fast leave-one-out AUC optimization on imbalanced prostate cancer data

Wang, G Teoh, JYC Lu, J

Choi, KS

Permalink

Publisher:: Springer Science and Business Media LLC
Publication Type:: Journal Article
Citation:: International Journal of Machine Learning and Cybernetics, 2020, 11, (8), pp. 1909-1922
Issue Date:: 2020-08-01

Closed Access

	Filename	Description	Size
	Wang2020_Article_LeastSquaresSupportVectorMachi.pdf	Published version	2.65 MB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Closed Access

This item is closed access and not available.

Full metadata record

Field	Value	Language
dc.contributor.author	Wang, G
dc.contributor.author	Teoh, JYC
dc.contributor.author	Lu, J https://orcid.org/0000-0003-0690-4732
dc.contributor.author	Choi, KS
dc.date.accessioned	2021-02-26T00:37:18Z
dc.date.available	2021-02-26T00:37:18Z
dc.date.issued	2020-08-01
dc.identifier.citation	International Journal of Machine Learning and Cybernetics, 2020, 11, (8), pp. 1909-1922
dc.identifier.issn	1868-8071
dc.identifier.issn	1868-808X
dc.identifier.uri	http://hdl.handle.net/10453/146475
dc.description.abstract	© 2020, Springer-Verlag GmbH Germany, part of Springer Nature. Quite often, the available pre-biopsy data for early prostate cancer detection are imbalanced. When the least squares support vector machines (LS-SVMs) are applied to such scenarios, it becomes naturally desirable for us to introduce the well-known AUC performance index into the LS-SVMs framework to avoid bias towards majority classes. However, this may result in high computational complexity for the minimal leave-one-out error. In this paper, by introducing the parameter λ, a generalized Area under the ROC curve (AUC) performance index RAUCLS is developed to theoretically guarantee that RAUCLS linearly depends on the classical AUC performance index RAUC. Based on both RAUCLS and the classical LS-SVM, a new AUC-based least squares support vector machine called AUC-LS-SVMs is proposed for directly and effectively classifying imbalanced prostate cancer data. The distinctive advantage of the proposed classifier AUC-LS-SVMs exists in that it can achieve the minimal leave-one-out error by quickly optimizing the parameter λ in RAUCLS using the proposed fast leave-one-out cross validation (LOOCV) strategy. The proposed classifier is first evaluated using generic public datasets. Further experiments are then conducted on a real-world prostate cancer dataset to demonstrate the efficacy of our proposed classifier for early prostate cancer detection.
dc.language	en
dc.publisher	Springer Science and Business Media LLC
dc.relation	http://purl.org/au-research/grants/arc/DP170101632
dc.relation.ispartof	International Journal of Machine Learning and Cybernetics
dc.relation.isbasedon	10.1007/s13042-020-01081-y
dc.rights	info:eu-repo/semantics/closedAccess
dc.subject	0801 Artificial Intelligence and Image Processing
dc.title	Least squares support vector machines with fast leave-one-out AUC optimization on imbalanced prostate cancer data
dc.type	Journal Article
utslib.citation.volume	11
utslib.for	0801 Artificial Intelligence and Image Processing
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
pubs.organisational-group	/University of Technology Sydney/Strength - AAII - Australian Artificial Intelligence Institute
pubs.organisational-group	/University of Technology Sydney
utslib.copyright.status	closed_access	*
dc.date.updated	2021-02-26T00:37:17Z
pubs.issue	8
pubs.publication-status	Published
pubs.volume	11
utslib.citation.issue	8

Abstract:

© 2020, Springer-Verlag GmbH Germany, part of Springer Nature. Quite often, the available pre-biopsy data for early prostate cancer detection are imbalanced. When the least squares support vector machines (LS-SVMs) are applied to such scenarios, it becomes naturally desirable for us to introduce the well-known AUC performance index into the LS-SVMs framework to avoid bias towards majority classes. However, this may result in high computational complexity for the minimal leave-one-out error. In this paper, by introducing the parameter λ, a generalized Area under the ROC curve (AUC) performance index RAUCLS is developed to theoretically guarantee that RAUCLS linearly depends on the classical AUC performance index RAUC. Based on both RAUCLS and the classical LS-SVM, a new AUC-based least squares support vector machine called AUC-LS-SVMs is proposed for directly and effectively classifying imbalanced prostate cancer data. The distinctive advantage of the proposed classifier AUC-LS-SVMs exists in that it can achieve the minimal leave-one-out error by quickly optimizing the parameter λ in RAUCLS using the proposed fast leave-one-out cross validation (LOOCV) strategy. The proposed classifier is first evaluated using generic public datasets. Further experiments are then conducted on a real-world prostate cancer dataset to demonstrate the efficacy of our proposed classifier for early prostate cancer detection.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/146475