Multivariate time series anomaly detection: A framework of Hidden Markov Models

• Published in: Applied soft computing Vol. 60; pp. 229 - 240
• Main Authors: Li, Jinbo, Pedrycz, Witold, Jamal, Iqbal
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.11.2017
• Subjects: Anomaly detection; Fuzzy C-means; Fuzzy integral; Hidden Markov Model (HMM); Multivariate time series; Multivariate time series; Hidden Markov Model (HMM); Fuzzy C-means; Fuzzy integral; Anomaly detection
• ISSN: 1568-4946; 1872-9681
• DOI: 10.1016/j.asoc.2017.06.035

[Display omitted] •Studied are multivariate time series anomaly detection.•Transformation methods are proposed.•They include clustering and hidden Markow model (HMM).•HMM detectors are constructed. In this study, we develop an approach to multivariate time series anomaly detection focused on the transformation of multivariate time series to univariate time series. Several transformation techniques involving Fuzzy C-Means (FCM) clustering and fuzzy integral are studied. In the sequel, a Hidden Markov Model (HMM), one of the commonly encountered statistical methods, is engaged here to detect anomalies in multivariate time series. We construct HMM-based anomaly detectors and in this context compare several transformation methods. A suite of experimental studies along with some comparative analysis is reported.