Real-Time Anomaly Detection in Edge Streams

TL;DR

This paper introduces MIDAS and MIDAS-F, real-time algorithms for detecting microcluster anomalies in dynamic graph streams, offering high accuracy, constant time processing, and theoretical false positive guarantees, outperforming existing methods.

Contribution

The paper presents MIDAS-F, an improved online anomaly detection method that mitigates poisoning effects and enhances accuracy for microcluster anomaly detection in graph streams.

Findings

MIDAS-F achieves significantly higher accuracy than MIDAS.

MIDAS provides theoretical false positive guarantees.

The algorithms process data in constant time and memory.

Abstract

Given a stream of graph edges from a dynamic graph, how can we assign anomaly scores to edges in an online manner, for the purpose of detecting unusual behavior, using constant time and memory? Existing approaches aim to detect individually surprising edges. In this work, we propose MIDAS, which focuses on detecting microcluster anomalies, or suddenly arriving groups of suspiciously similar edges, such as lockstep behavior, including denial of service attacks in network traffic data. We further propose MIDAS-F, to solve the problem by which anomalies are incorporated into the algorithm's internal states, creating a `poisoning' effect that can allow future anomalies to slip through undetected. MIDAS-F introduces two modifications: 1) We modify the anomaly scoring function, aiming to reduce the `poisoning' effect of newly arriving edges; 2) We introduce a conditional merge step, which updates the algorithm's data structures after each time tick, but only if the anomaly score is below a threshold value, also to reduce the `poisoning' effect. Experiments show that MIDAS-F has significantly higher accuracy than MIDAS. MIDAS has the following properties: (a) it detects microcluster anomalies while providing theoretical guarantees about its false positive probability; (b) it is online, thus processing each edge in constant time and constant memory, and also processes the data orders-of-magnitude faster than state-of-the-art approaches; (c) it provides up to 62% higher ROC-AUC than state-of-the-art approaches.