PIF: Anomaly detection via preference embedding

TL;DR

The paper introduces PIF, a novel anomaly detection method that combines adaptive isolation with preference embedding, using a high-dimensional embedding and PI-Forest for efficient scoring, outperforming existing techniques.

Contribution

It presents PIF, a new anomaly detection approach that leverages preference embedding and a tree-based PI-Forest for improved performance and flexibility.

Findings

PIF outperforms state-of-the-art anomaly detection methods.

PI-Forest effectively measures arbitrary distances.

PIF successfully isolates anomalies in preference space.

Abstract

We address the problem of detecting anomalies with respect to structured patterns. To this end, we conceive a novel anomaly detection method called PIF, that combines the advantages of adaptive isolation methods with the flexibility of preference embedding. Specifically, we propose to embed the data in a high dimensional space where an efficient tree-based method, PI-Forest, is employed to compute an anomaly score. Experiments on synthetic and real datasets demonstrate that PIF favorably compares with state-of-the-art anomaly detection techniques, and confirm that PI-Forest is better at measuring arbitrary distances and isolate points in the preference space.