Previous [ 1] [ 2] [ 3] [ 4] [ 5] [ 6] [ 7] [ 8] [ 9] [ 10] [ 11] [ 12] [ 13] [ 14] [ 15] [ 16] [ 17] [ 18] [ 19]

@

Journal of Information Science and Engineering, Vol. 30 No. 6, pp. 1985-2002 (November 2014)


Support Vector Domain Description with Maximum Between Spheres Separability


MOHAMED EL BOUJNOUNI1, MOHAMED JEDRA2 AND NOUREDDINE ZAHID3
Faculty of Sciences
Laboratory of Conception and Systems (Microelectronic and Informatics)
Mohammed v V Agdal University
B.P. 1014, Rabat, Morocco
E-mail: med_elbouj@yahoo.fr1; jedra@fsr.ac.ma2; zahid@fsr.ac.ma3

Support Vector Domain Description (SVDD) is inspired by the Support Vector Classifier. It obtains a sphere shaped decision boundary with minimal volume around a dataset. This data description can be used for novelty or outlier detection. Our approach is always to minimize the volume of the sphere describing the dataset, while at the same time maximize the separability between the spheres. To build such sphere we only need to solve a convex optimization problem that can be efficiently solved with the existing software packages, simulation results on seventeen benchmark datasets have successfully validated the effectiveness of the proposed method.

Keywords: support vector domain description, small hypersphere, maximum separability

Full Text () Retrieve PDF document (201411_17.pdf)

Received October 28, 2012; revised May 20 & October 28, 2013; accepted December 19, 2013.
Communicated by Hsin-Min Wang.