An outlier detection algorithm based on a soft hyper-sphere for high dimension nonlinear data

XU Gang; ZHANG Xiao-tong; LI Min; XU Jin-wu

doi:10.13374/j.issn2095-9389.2017.10.014

Volume 39 Issue 10

Oct. 2017

Turn off MathJax

Article Contents

Article Navigation > Chinese Journal of Engineering > 2017 > 39(10): 1552-1558

XU Gang, ZHANG Xiao-tong, LI Min, XU Jin-wu. An outlier detection algorithm based on a soft hyper-sphere for high dimension nonlinear data[J]. Chinese Journal of Engineering, 2017, 39(10): 1552-1558. doi: 10.13374/j.issn2095-9389.2017.10.014

Citation:

XU Gang, ZHANG Xiao-tong, LI Min, XU Jin-wu. An outlier detection algorithm based on a soft hyper-sphere for high dimension nonlinear data[J]. Chinese Journal of Engineering, 2017, 39(10): 1552-1558. doi: 10.13374/j.issn2095-9389.2017.10.014

Citation:

PDF( 2323 KB)

An outlier detection algorithm based on a soft hyper-sphere for high dimension nonlinear data

doi: 10.13374/j.issn2095-9389.2017.10.014

XU Gang^{1,2
,
,},
ZHANG Xiao-tong¹,
LI Min²,
XU Jin-wu²

1) School of Computer and Communication Engineering, University of Science and Technology Beijing, Beijing 100083, China
2) Collaborative Innovation Center of Steel Technology, Beijing 100083, China

Received Date: 2016-07-07

Abstract

Abstract

In process industries, such as metallurgy and chemistry, real procedure parameters usually possess high-dimensional nonlinear features. To solve the problem of outlier detection in complex high-dimensional data, the concept of a soft hyper-sphere is introduced in this paper. An original data set is projected into a high-dimensional feature space using a nonlinear kernel function, and the boundary of the soft hyper-sphere is determined within this feature space. To avoid a mass product quality incident, location information on the testing samples, which are projected into the feature space, is used to decide whether they are outliers. As an applied example, practical procedure data obtained from a type of auto steel product were tested. The results verify that the proposed outlier detection algorithm based on a soft hyper-sphere has a better ability for outlier detection in high-dimensional nonlinear data than tradional methods.
- soft hyper-sphere,
- high dimension nonlinear data,
- outlier detection algorithm,
- Kernel mapping

FullText(HTML)

References(10)

References

[1]	Zimek A, Schubert E, Kriegel H P. A survey on unsupervised outlier detection in high-dimensional numerical data. Statist Anal Data Min ASA Data Sci, 2012, 5(5):363
[2]	Guo J H, Huang W, Williams B M. Real time traffic flow outlier detection using short-term traffic conditional variance prediction. Transport Res C Emerg Technol, 2015, 50:160
[6]	Johnson J E. A User's Guide to Principal Components. New York:John Wiley&Sons Inc, 1991
[7]	Wold S, Martens H, Wold H. The multivariate calibration problem in chemistry solved by the PLS method. Matrix Pencils, 1983, 973:286
[8]	Tax D M J. One-Class Classification[Dissertation]. Dutch:Delft University of Technology, 2001
[9]	Liu B, Xiao Y, Cao L, et al. SVDD-based outlier detection on uncertain data. Knowl Inform Syst, 2013, 34(3):597
[10]	Sakla W, Chan A, Ji J, et al. An SVDD-based algorithm for target detection in hyperspectral imagery. IEEE Geosci Remote Sens Lett, 2011, 8(2):384
[11]	Shawe-Taylor J, Cristianini N. Kernel Methods for Pattern Analysis. England:Cambridge University Press, 2004
[12]	Rosipal R, Trejo L J. Kernel partial least squares regression in reproducing Kernel Hillbert space. J Mach Learn Res, 2001, 2:97
[13]	Bach F R, Jordan M I. Kernel independent component analysis. J Mach Learn Res, 2002, 3:1