Learning Invariant Rules from Data for Interpretable Anomaly Detection

TL;DR

This paper introduces a framework that learns invariant rules from data to improve the interpretability of anomaly detection, providing explicit explanations for anomalies while maintaining competitive detection performance.

Contribution

It proposes a novel method to automatically learn invariant rules that explain anomalies, enhancing interpretability without sacrificing detection accuracy.

Findings

Achieves comparable or better AUC and partial AUC on benchmark datasets.

Provides explicit explanations for detected anomalies.

Demonstrates effectiveness across various application domains.

Abstract

In the research area of anomaly detection, novel and promising methods are frequently developed. However, most existing studies exclusively focus on the detection task only and ignore the interpretability of the underlying models as well as their detection results. Nevertheless, anomaly interpretation, which aims to provide explanation of why specific data instances are identified as anomalies, is an equally important task in many real-world applications. In this work, we propose a novel framework which synergizes several machine learning and data mining techniques to automatically learn invariant rules that are consistently satisfied in a given dataset. The learned invariant rules can provide explicit explanation of anomaly detection results in the inference phase and thus are extremely useful for subsequent decision-making regarding reported anomalies. Furthermore, our empirical evaluation shows that the proposed method can also achieve comparable or even better performance in terms of AUC and partial AUC on public benchmark datasets across various application domains compared with start-of-the-art anomaly detection models.