Maximum margin novelty detection method based on small amount of abnormal data
基于少量异常数据的最大间隔新奇检测方法*

Traditional novel detection algorithms often only use the normal samples which account for most of the total sample to construct a classifier, the negative class samples are ineffective. To solve this problem, this paper proposes a large margin method that based on a small amount of abnormal data (BSLM). The basic idea is as below: First, a hypersphere should be constructed to contain as many normal instances as possible, and at the same time make sure the margin between the surface of this sphere and normal instances are large enough. In this way, a closed boundary can be attained which surrounds the normal data and is tightly close to the abnormal data. To build such a sphere we only need to solve a convex optimization problem, which can be efficiently solved through the existing traditional support vector machine (SVM) model with a little change. By some simulation experiments on the datasets of the machine fault detection, medical diagnosis, and the Arabic numeral recognition, the results show that this method can effectively improve the true positive rate, reduce the false positive rate. At the same time fivefold cross-validation training methods increases the detection stability.