Real-Time Anomaly Detection with Synthetic Anomaly Monitoring (SAM)

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Luzio, Emanuele, Ponti, Moacir Antonelli
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915134120656896
author Luzio, Emanuele
Ponti, Moacir Antonelli
author_facet Luzio, Emanuele
Ponti, Moacir Antonelli
contents Anomaly detection is essential for identifying rare and significant events across diverse domains such as finance, cybersecurity, and network monitoring. This paper presents Synthetic Anomaly Monitoring (SAM), an innovative approach that applies synthetic control methods from causal inference to improve both the accuracy and interpretability of anomaly detection processes. By modeling normal behavior through the treatment of each feature as a control unit, SAM identifies anomalies as deviations within this causal framework. We conducted extensive experiments comparing SAM with established benchmark models, including Isolation Forest, Local Outlier Factor (LOF), k-Nearest Neighbors (kNN), and One-Class Support Vector Machine (SVM), across five diverse datasets, including Credit Card Fraud, HTTP Dataset CSIC 2010, and KDD Cup 1999, among others. Our results demonstrate that SAM consistently delivers robust performance, highlighting its potential as a powerful tool for real-time anomaly detection in dynamic and complex environments.
format Preprint
id arxiv_https___arxiv_org_abs_2501_18417
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Real-Time Anomaly Detection with Synthetic Anomaly Monitoring (SAM)
Luzio, Emanuele
Ponti, Moacir Antonelli
Machine Learning
62H30, 68T05, 62G99, 91G80, 68M10
I.5.4; K.6.5; I.2.6; H.4.2
Anomaly detection is essential for identifying rare and significant events across diverse domains such as finance, cybersecurity, and network monitoring. This paper presents Synthetic Anomaly Monitoring (SAM), an innovative approach that applies synthetic control methods from causal inference to improve both the accuracy and interpretability of anomaly detection processes. By modeling normal behavior through the treatment of each feature as a control unit, SAM identifies anomalies as deviations within this causal framework. We conducted extensive experiments comparing SAM with established benchmark models, including Isolation Forest, Local Outlier Factor (LOF), k-Nearest Neighbors (kNN), and One-Class Support Vector Machine (SVM), across five diverse datasets, including Credit Card Fraud, HTTP Dataset CSIC 2010, and KDD Cup 1999, among others. Our results demonstrate that SAM consistently delivers robust performance, highlighting its potential as a powerful tool for real-time anomaly detection in dynamic and complex environments.
title Real-Time Anomaly Detection with Synthetic Anomaly Monitoring (SAM)
topic Machine Learning
62H30, 68T05, 62G99, 91G80, 68M10
I.5.4; K.6.5; I.2.6; H.4.2
url https://arxiv.org/abs/2501.18417