Robust anomaly and change detection utilizing sparse decomposition

Abstract

The invention relates to robust anomaly and change detection utilizing sparse decomposition. The present disclosure describes systems, non-transitory computer-readable media, and methods for determining latent components of a metrics time series and identifying anomalous data within the metrics time series based on one or both of spikes / dips and level changes from the latent components satisfying significance thresholds. To identify such latent components, in some cases, the disclosed systems account for a range of value types by intelligently subjecting real values to a latent-component constraint for decomposing the time series and intelligently excluding non-real values from the latent-component constraint. The disclosed systems can further identify significant anomalous data values from latent components of the metrics time series by jointly determining whether one or both of a subseries of a spike-component series and a level change from a level-component series satisfy significance thresholds.

Description

Background technique [0001] In recent years, analytical computing systems have improved the accuracy of identifying trends and seasonal changes by deploying new algorithms for analyzing time series, including datasets of metrics recorded across time. For example, a conventional analytical computing system may identify and present anomalies from a time series representing user actions with respect to the operation of a website, web-accessible application, or other web-based device. To illustrate, some existing systems may split time series of large amounts of network metric data into components as a basis for identifying anomalous metrics in the time series, such as unusual user actions outside expected trends. [0002] Although conventional analytical computing systems can identify anomalies in time series, such systems may inaccurately and ineffectively identify outliers within the time series by applying conventional anomaly detection algorithms. For example, conventional s...

AI Technical Summary

Problems solved by technology

However, strict reliance on test and training periods for data within a time series ignores important changes in the time series that can lead to key analytical insights

Images