Exponentially Weighted Moving Average Charts for Detecting Concept Drift

TL;DR

This paper introduces a computationally efficient EWMA-based method for detecting concept drift in streaming data, which can be integrated with any classifier and controls false positive rates.

Contribution

It presents a novel, modular, and online EWMA chart approach for concept drift detection that is adaptable and maintains a constant false positive rate.

Findings

Method is computationally efficient with O(1) overhead.

Works in a fully online manner without data storage.

Allows control of false positive detection rate.

Abstract

Classifying streaming data requires the development of methods which are computationally efficient and able to cope with changes in the underlying distribution of the stream, a phenomenon known in the literature as concept drift. We propose a new method for detecting concept drift which uses an Exponentially Weighted Moving Average (EWMA) chart to monitor the misclassification rate of an streaming classifier. Our approach is modular and can hence be run in parallel with any underlying classifier to provide an additional layer of concept drift detection. Moreover our method is computationally efficient with overhead O(1) and works in a fully online manner with no need to store data points in memory. Unlike many existing approaches to concept drift detection, our method allows the rate of false positive detections to be controlled and kept constant over time.