Classification Framework for Anomaly Detection

摘要

One way to describe anomalies is by saying that anomalies are not concentrated. This leads to the problem of .nding level sets for the data generating density. We interpret this learning problem as a binary classi.cation problem and compare the corresponding classi.cation risk with the standard performance measure for the density level problem. In particular it turns out that the empirical classi.cation risk can serve as an empirical performance measure for the anomaly detection problem. This allows us to compare di.erent anomaly detection algorithms empirically, i.e. with the help of a test set. Based on the above interpretation we then propose a support vector machine (SVM) for anomaly detection. Finally, we establish universal consistency for this SVM and report some experiments which compare our SVM to other commonly used methods including the standard one-class SVM.