Real Time Anomaly Detection And Categorisation

DOI10.48550/ARXIV.2009.06670zbMATH Open1490.62014arXiv2009.06670OpenAlexW3086253642MaRDI QIDQ119893FDOQ119893

Authors: Alexander T. M. Fisch, Lawrence Bardwell, Idris Eckley, Alexander T. M. Fisch, Lawrence Bardwell, Idris A. Eckley 

Publication date: 14 September 2020

Published in: Statistics and Computing (Search for Journal in Brave)

Abstract: The ability to quickly and accurately detect anomalous structure within data sequences is an inference challenge of growing importance. This work extends recently proposed post-hoc (offline) anomaly detection methodology to the sequential setting. The resultant procedure is capable of real-time analysis and categorisation between baseline and two forms of anomalous structure: point and collective anomalies. Various theoretical properties of the procedure are derived. These, together with an extensive simulation study, highlight that the average run length to false alarm and the average detection delay of the proposed online algorithm are very close to that of the offline version. Experiments on simulated and real data are provided to demonstrate the benefits of the proposed method.

Full work available at URL: https://arxiv.org/abs/2009.06670

Recommendations

• A real-time system-adapted anomaly detector
• An adaptive approach to granular real-time anomaly detection
• A classification framework for anomaly detection
• Anomaly Detection in Streaming Nonstationary Temporal Data
• scientific article; zbMATH DE number 2089935

zbMATH Keywords

anomaly detectionreal timeSCAPAstreaming data

Mathematics Subject Classification ID

Computational methods for problems pertaining to statistics (62-08) Time series, auto-correlation, regression, etc. in statistics (GARCH) (62M10) Applications of statistics to biology and medical sciences; meta analysis (62P10)

Cites Work

• A MOSUM procedure for the estimation of multiple random change points
• Robustness properties of \(S\)-estimators of multivariate location and shape in high dimension
• Optimal detection of changepoints with a linear computational cost
• Changepoint Detection in the Presence of Outliers
• Circular binary segmentation for the analysis of array-based DNA copy number data
• Simultaneous discovery of rare and common segment variants
• Procedures for Reacting to a Change in Distribution
• CONTINUOUS INSPECTION SCHEMES
• Optimal detection of a change in distribution
• Bayesian detection of abnormal segments in multiple time series
• Robust Kalman tracking and smoothing with propagating and non-propagating outliers
• Evaluating stationarity via change-point alternatives with applications to fMRI data
• Efficient on-line estimation of autoregressive parameters
• A two-step sequential procedure for detecting an epidemic change
• A Space-Efficient Recursive Procedure for Estimating a Quantile of an Unknown Distribution
• Joint Estimation of Model Parameters and Outlier Effects in Time Series
• Tests for change-points with epidemic alternatives
• Subset Multivariate Collective and Point Anomaly Detection

Cited In (8)

• Online anomaly detection of profiles with varying coefficients via functional mixed effects modelling
• Anomaly detection of mobile positioning data with applications to COVID-19 situational awareness
• Inductive conformal anomaly detection for sequential detection of anomalous sub-trajectories
• Unsupervised streaming anomaly detection for instrumented infrastructure
• Iterative selection of categorical variables for log data anomaly detection
• Collective Anomaly Detection in High-Dimensional Var Models
• Anomaly detection for online visiting traffic as a real-estate indicator: the case of homebuyer
• Bayesian root cause analysis by separable likelihoods

Uses Software

• robust
• mosum

This page was built for publication: Real Time Anomaly Detection And Categorisation

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q119893)