article

Exponentially weighted moving average charts for detecting concept drift

Authors:

Gordon J. Ross,

Niall M. Adams,

Dimitris K. Tasoulis,

David J. HandAuthors Info & Claims

Pattern Recognition Letters, Volume 33, Issue 2

Pages 191 - 198

https://doi.org/10.1016/j.patrec.2011.08.019

Published: 01 January 2012 Publication History

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.

References

[1]

Paired learners for concept drift. In: ICDM'08: Proc. 2008 Eighth IEEE Internat. Conf. on Data Mining, IEEE Computer Society, Washington, DC, USA. pp. 23-32.

[2]

Baena-Garcia, M., Campo-Avila, J., Fidalgo, R., 2006. Early drift detection. In: Advances in Aritificial Intelligence, Lecture Notes Artificial Intelligence, vol. 3171, pp. 286-295.

[3]

Detection of Abrupt Change Theory and Application. Prentice Hall.

[4]

Bell, C., Gordon, L., Pollak, M., 1994. An efficient nonparametric detection scheme and its application to surveillance of a Bernoulli process with unknown baseline. In: Change Point Problem, vol. 23, pp. 7-27.

[5]

Approximate statistical tests for comparing supervised classification learning algorithms. Neural Comput. v10. 1895-1923.

[6]

A general framework for mining massive data stream. J. Comput. Graph. Statist. v12. 2003

[7]

Gama, J., Medas, P., Castillo, G., Rodrigues, P., 2004. Learning with drift detection. In: Advances in Artificial Intelligence, Lecture Notes in Artificial Intelligence, vol. 3171, pp. 286-295.

[8]

Statistical and structural approaches to texture. Proc. IEEE. v67 i5. 786-804.

[9]

Harries, M., 1999. Splice-2 comparative evaluation: Electricity pricing. Tech. rep., The University of New South Wales.

[10]

The Elements of Statistical Learning. Springer.

[11]

Image recognition and neuronal networks: Intelligent systems for the improvement of imaging information. Minimal Invasive Therapy Allied Technol. v9 i3-4. 225-230.

[12]

Klinkenberg, R., Joachims, T., 2000. Detecting concept drift with support vector machines. In: Proc. 17th Internat. Conf. on Machine Learning. ICML '00, pp. 487-494.

[13]

Dynamic weighted majority: An ensemble method for driffting concepts. J. Machine Learn. Res. v8. 2755-2790.

[14]

Kuncheva, L.I., 2009. Using control charts for detecting concept change in streaming data. Tech. rep., Bangor University.

[15]

Kuncheva, L.I., Plumpton, C.O., 2008. Adaptive learning rate for online linear discriminant classifiers. In: Structural, Syntactic, and Statistical Pattern Recognition, Lecture Notes in Computer Science, vol. 5342, pp. 510-519.

[16]

Law, Y.-N., Zaniolo, C., 2005. An adaptive nearest neighbor classification algorithm for data streams. In: PKDD, pp. 108-120.

[17]

A simple cumulative sum type statistic for the change-point problem with zero-one observations. Biometrika. v67 i1. 79-84.

[18]

A CUSUM chart for monitoring a proportion when inspecting continuously. J. Qual. Technol. v31 i1. 87-108.

[19]

Control chart tests based on geometric moving averages. Technometrics. v42 i1. 97-101.

[20]

CUSUM charts for signalling varying location shifts. J. Qual. Technol. v32.

[21]

Adaptive threshold computation for CUSUM-type procedures in change detection and isolation problems. Comput. Statist. Data Anal. v52 i9. 4161-4174.

[22]

A comparison of state-of-the-art classification techniques for expert automobile insurance claim fraud detection. J. Risk Insurance. v69. 373-421.

[23]

Learning in the presence of concept drift and hidden contexts. Machine Learn. v23. 69-101.

[24]

EWMA control charts for monitoring high-yield processes based on non-transformed observations. Internat. J. Prod. Res. v46 i20. 5679-5699.

Cited By

Mahdi OAli NPardede EAl-Quraishi T(2024)Online Concept Drift DetectorProcedia Computer Science10.1016/j.procs.2024.05.140237:C(559-567)Online publication date: 24-Jul-2024
https://dl.acm.org/doi/10.1016/j.procs.2024.05.140
Vandervorst FVerbeke WVerdonck T(2024)Claims fraud detection with uncertain labelsAdvances in Data Analysis and Classification10.1007/s11634-023-00568-018:1(219-243)Online publication date: 1-Mar-2024
https://dl.acm.org/doi/10.1007/s11634-023-00568-0
Arora SRani RSaxena N(2024)SETL: a transfer learning based dynamic ensemble classifier for concept drift detection in streaming dataCluster Computing10.1007/s10586-023-04149-w27:3(3417-3432)Online publication date: 1-Jun-2024
https://dl.acm.org/doi/10.1007/s10586-023-04149-w
Show More Cited By

Exponentially weighted moving average charts for detecting concept drift
1. Computing methodologies
  1. Machine learning
    1. Learning paradigms
      1. Supervised learning
    2. Machine learning approaches

Recommendations

Detecting group concept drift from multiple data streams
Highlights
- Proposing a new type of concept drift group concept drift that commonly exists in multiple data streams.
Abstract
Concept drift may lead to a sharp downturn in the performance of streaming in data-based algorithms, caused by unforeseeable changes in the underlying distribution of data. In this paper, we are mainly concerned with concept drift ...
Concept drift detection and localization framework based on behavior replacement
Abstract
With the application of numerous services or software, process mining has attracted more and more attention. However, concept drift may occur during process mining due to the instability of the process. Sudden and gradual drifts are considered to ...
Detecting concept drift in data streams using model explanation

A novel concept drift detector for data streams is proposed.The drift detector can be combined with an arbitrary classification algorithm.The drift detector uses model explanation to detect concept drift.The approach features good drift detection, ...

Comments

Please enable JavaScript to view thecomments powered by Disqus.

Information & Contributors

Information

Published In

cover image Pattern Recognition Letters

Pattern Recognition Letters Volume 33, Issue 2

January, 2012

125 pages

ISSN:0167-8655

Issue’s Table of Contents

Copyright © © 2011.

Publisher

Elsevier Science Inc.

United States

Publication History

Published: 01 January 2012

Author Tags

Qualifiers

Article

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

57
Total Citations
View Citations
1
Total Downloads

Downloads (Last 12 months)0
Downloads (Last 6 weeks)0

Reflects downloads up to 13 Nov 2024

Other Metrics

View Author Metrics

Citations

Cited By

Mahdi OAli NPardede EAl-Quraishi T(2024)Online Concept Drift DetectorProcedia Computer Science10.1016/j.procs.2024.05.140237:C(559-567)Online publication date: 24-Jul-2024
https://dl.acm.org/doi/10.1016/j.procs.2024.05.140
Vandervorst FVerbeke WVerdonck T(2024)Claims fraud detection with uncertain labelsAdvances in Data Analysis and Classification10.1007/s11634-023-00568-018:1(219-243)Online publication date: 1-Mar-2024
https://dl.acm.org/doi/10.1007/s11634-023-00568-0
Arora SRani RSaxena N(2024)SETL: a transfer learning based dynamic ensemble classifier for concept drift detection in streaming dataCluster Computing10.1007/s10586-023-04149-w27:3(3417-3432)Online publication date: 1-Jun-2024
https://dl.acm.org/doi/10.1007/s10586-023-04149-w
Davalas CMichail DDiou CVarlamis ITserpes K(2024)A rehearsal framework for computational efficiency in online continual learningApplied Intelligence10.1007/s10489-024-05493-z54:8(6383-6399)Online publication date: 1-Apr-2024
https://dl.acm.org/doi/10.1007/s10489-024-05493-z
Angbera AChan H(2024)SABeDM: a sliding adaptive beta distribution model for concept drift detection in a dynamic environmentKnowledge and Information Systems10.1007/s10115-023-02004-366:3(2039-2062)Online publication date: 1-Mar-2024
https://dl.acm.org/doi/10.1007/s10115-023-02004-3
Ashok SEzhumalai SPatwa T(2023)Remediating data drifts and re-establishing ML modelsProcedia Computer Science10.1016/j.procs.2023.01.060218:C(799-809)Online publication date: 1-Jan-2023
https://dl.acm.org/doi/10.1016/j.procs.2023.01.060
Komorniczak JKsieniewicz P(2023)Complexity-based drift detection for nonstationary data streamsNeurocomputing10.1016/j.neucom.2023.126554552:COnline publication date: 1-Oct-2023
https://dl.acm.org/doi/10.1016/j.neucom.2023.126554
Jin CFeng YLi F(2023)Concept drift detection based on decision distribution in inconsistent information systemKnowledge-Based Systems10.1016/j.knosys.2023.110934279:COnline publication date: 4-Nov-2023
https://dl.acm.org/doi/10.1016/j.knosys.2023.110934
Tan CLee VSalehi M(2022)Information resources estimation for accurate distribution-based concept drift detectionInformation Processing and Management: an International Journal10.1016/j.ipm.2022.10291159:3Online publication date: 1-May-2022
https://dl.acm.org/doi/10.1016/j.ipm.2022.102911
Yu HZhang QLiu TLu JWen YZhang G(2022)Meta-ADDInformation Sciences: an International Journal10.1016/j.ins.2022.07.022608:C(996-1009)Online publication date: 1-Aug-2022
https://dl.acm.org/doi/10.1016/j.ins.2022.07.022
Show More Cited By

View Options

View options

Get Access

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Full Access

Get this Publication

Media

Figures

Other

Tables

View Issue’s Table of Contents