Automata Based Multivariate Time Series Analysis for Anomaly Detection over Sliding Time Windows

Abstract

We describe an optimal linear time complexity method for extracting patterns from sliding windows of multivariate time series that depends only on the length of the time series. The method is implemented as an open-source Java library and is used to detect anomalies in multivariate time series.

Keywords

• multivariate time series
• transducers
• sliding windows
• anomaly detection