OHODIN – Online Anomaly Detection for Data Streams

Abstract

We propose OHODIN an online extension for data streams of the knn-based ODIN anomaly detection approach and presents a detection-threshold heuristic that is based on extreme value theory. In contrast to sophisticated anomaly and novelty detection approaches the decision-making process of ODIN is interpretable by humans, making it interesting for certain applications. This article presents the algorithms itself and an experimental evaluation with competing state-of-the-art anomaly detection approaches.