Kernel ellipsoidal trimming
Kernel ellipsoidal trimming
Ellipsoid estimation is important in many practical areas such as control, system identification, visual/audio tracking, experimental design, data mining, robust statistics and statistical outlier or novelty detection. A new method, called kernel minimum volume covering ellipsoid (KMVCE) estimation, that finds an ellipsoid in a kernel-defined feature space is presented. Although the method is very general and can be applied to many of the aforementioned problems, the main focus is on the problem of statistical novelty/outlier detection. A simple iterative algorithm based on Mahalanobis-type distances in the kernel-defined feature space is proposed for practical implementation. The probability that a non-outlier is misidentified by our algorithms is analyzed using bounds based on Rademacher complexity. The KMVCE method performs very well on a set of real-life and simulated datasets, when compared with standard kernel-based novelty detection methods.
minimum volume covering ellipsoid, rademacher complexity, kernel methods, outlier detection, novelty detection
309-324
Dolia, A.N.
5bf0ed58-7341-4147-8d05-8bab99a6d038
Harris, C.J.
c4fd3763-7b3f-4db1-9ca3-5501080f797a
Shawe-Taylor, J.S.
455c50d6-e793-4695-8808-ee67a1d29e0b
Titterington, D.M.
bdfdf43b-7423-4842-9eab-740a5b840e24
2007
Dolia, A.N.
5bf0ed58-7341-4147-8d05-8bab99a6d038
Harris, C.J.
c4fd3763-7b3f-4db1-9ca3-5501080f797a
Shawe-Taylor, J.S.
455c50d6-e793-4695-8808-ee67a1d29e0b
Titterington, D.M.
bdfdf43b-7423-4842-9eab-740a5b840e24
Dolia, A.N., Harris, C.J., Shawe-Taylor, J.S. and Titterington, D.M.
(2007)
Kernel ellipsoidal trimming.
Computational Statistics and Data Analysis, 52 (1), .
(doi:10.1016/j.csda.2007.03.020).
Abstract
Ellipsoid estimation is important in many practical areas such as control, system identification, visual/audio tracking, experimental design, data mining, robust statistics and statistical outlier or novelty detection. A new method, called kernel minimum volume covering ellipsoid (KMVCE) estimation, that finds an ellipsoid in a kernel-defined feature space is presented. Although the method is very general and can be applied to many of the aforementioned problems, the main focus is on the problem of statistical novelty/outlier detection. A simple iterative algorithm based on Mahalanobis-type distances in the kernel-defined feature space is proposed for practical implementation. The probability that a non-outlier is misidentified by our algorithms is analyzed using bounds based on Rademacher complexity. The KMVCE method performs very well on a set of real-life and simulated datasets, when compared with standard kernel-based novelty detection methods.
This record has no associated files available for download.
More information
Published date: 2007
Keywords:
minimum volume covering ellipsoid, rademacher complexity, kernel methods, outlier detection, novelty detection
Organisations:
Southampton Statistical Research Inst.
Identifiers
Local EPrints ID: 48104
URI: http://eprints.soton.ac.uk/id/eprint/48104
ISSN: 0167-9473
PURE UUID: d5818f29-da7c-41af-adf0-808c268c2531
Catalogue record
Date deposited: 28 Aug 2007
Last modified: 15 Mar 2024 09:43
Export record
Altmetrics
Contributors
Author:
A.N. Dolia
Author:
C.J. Harris
Author:
J.S. Shawe-Taylor
Author:
D.M. Titterington
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics