Density Level Detection is Classification

We show that anomaly detection can be interpreted as a binary classification problem. Using this interpretation we propose a support vector machine (SVM) for anomaly detection. We then present some theoretical results which include consistency and learning rates. Finally, we experimentally compare our SVM with the standard one-class SVM.

• Type: Conference
• Year: 2004
• Where: NIPS
• Authors: Ingo Steinwart, Don R. Hush, Clint Scovel
• Keywords: Anomaly Detection, Binary Classification Problem, NIPS 2004, NIPS 2007, Standard One-class Svm