New one-class classifiers based on the origin separation approach
Mario Michael Krell, Hendrik Wöhrle
In Pattern Recognition Letters, Elsevier, volume 53, pages 93-99, Feb/2015.

Abstract :

The model of the one-class support vector machine (voc-SVM) is based on the "origin separation approach", i.e., to add a sample at the origin to the training data for the second class and apply a maximum margin separation as known from the C-SVM. This has been proven only for hard margin separation but a clearly defined relation between the voc- SVM and the classical SVM (C-SVM) is not yet existing. In this work, the origin separation approach is analyzed in more detail. The approach reveals to be a more general concept to relate binary and unary (one-class) classifiers. We prove how its application to the v-SVM, a variant of the C-SVM, directly results in the voc-SVM. Furthermore, we apply this concept to the C-SVM and other related methods (balanced relative margin machine, regularized Fisher's discriminant analysis, online passive-aggressive algorithms) to derive entirely new classifiers. This includes variants that can be updated online which allows the application on large datasets or on systems with very limited resources.


last updated 06.09.2016
to top