A Brief Survey on Concept Drift

Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 308)


The digital universe is growing rapidly. The volume of data generated per annum is in the order of zeta bytes due to the proliferation of the Internet. Many real-world applications generate data that are continuous. This type of data is known as data streams. Examples of applications generating this kind of data are business transactions, Web logs, sensors networks, etc. The data stream is analyzed, and the underlying concepts are extracted to make predictions and decisions in real time. But as data streams evolve over time, they undergo concept drift. Concept drift means the statistical properties of the data stream change over time in unforeseen ways. This causes problems because the predictions based on the data streams become less accurate as time passes. To understand the behavior of data streams, it is important to investigate the changes of the data distributions and the causes of the changes. Therefore, periodic retraining, also known as refreshing, of any model is necessary. The survey covers the various techniques available in the literature to handle concept drift in data streams.


Data distributions Data stream Concept drift 


  1. 1.
    Tsymbal, A.: The problem of concept drift: definitions and related work. Technical report TCD-CS-2004-15, Trinity College Dublin, Ireland, pp. 1–7 (2004)Google Scholar
  2. 2.
    Masud, M., Chen, Q., Khan, L., Aggarwal, C., Gao, J., Han, J., Thuraisingham, B.: Addressing concept-evolution in concept-drifting data streams. IEEE International Conference on Data Mining, pp. 929–934 (2010)Google Scholar
  3. 3.
    Alippi, C., Boracchi, G.,Roveri, M.: An effective just-in-time adaptive classifier for gradual concept drifts. In: International Joint Conference on Neural Networks, pp. 1675–1682 (2011)Google Scholar
  4. 4.
    Brzezinski, D., Stefanowski, J.: Accuracy updated ensemble for data streams with concept drift. In: 6th International Conference on Hybrid Artificial Intelligent Systems. Lecture Notes in Computer Science, vol. 6679, pp. 155–162. Springer (2011)Google Scholar
  5. 5.
    Shetty, S., Mukkavilli, S.K., Keel, L.H.: An integrated machine learning and control theoretic model for mining concept-drifting data streams. In: IEEE International Conference on Technologies for Homeland Security (HST), pp. 75–80 (2011)Google Scholar
  6. 6.
    Breve, F., Zhao, L.: Particle competition and cooperation in networks for semi-supervised learning with concept drift. In: The 2012 International Joint Conference on Neural Networks (IJCNN), pp. 1–6 (2012)Google Scholar
  7. 7.
    Bifet, A., Read, J., Pfahringer, B., Holmes, G.: Efficient data stream classification via probabilistic adaptive windows. In: Proceedings of the 28th Annual ACM Symposium on Applied Computing, pp. 801–806 (2013)Google Scholar
  8. 8.
    Turkov, P., Krasotkina, O., Mottl, V.: The bayesian logistic regression in pattern recognition problems under concept drift. In: International Conference on Pattern Recognition, pp. 2976–2979 (2012)Google Scholar
  9. 9.
    Minku, L.L., Yao, X.: DDD: New ensemble approach for dealing with concept drift. IEEE Trans. Knowl. Data Eng. 24(4), 619–633 (2012)CrossRefGoogle Scholar
  10. 10.
    Gama, J., Medas, P., Castillo, G., Pedro Rodrigues, P.: Learning with drift detection. In: Advances in Artificial Intelligence, pp. 286–295. Springer, Berlin Heidelberg (2004)Google Scholar
  11. 11.
    Ross, G.J., Adams, N.M., Tasoulis, D.K., Hand, D.J.: Exponentially weighted moving average charts for detecting concept drift. Pattern Recognit. Lett. 33(2), 191–198 (Elsiever) (2012)Google Scholar
  12. 12.
    Brzezinski, D., Stefanowski, J.: Reacting to different types of concept drift: the accuracy updated ensemble algorithm. IEEE Trans. Neural Netw. Learn. Syst. 25(1), 81–94 (IEEE) (2013)Google Scholar
  13. 13.
    Susnjak, T., Barczak, A.L.C., Hawick, K.A.: Adaptive ensemble based learning in non-stationary environments. In: International Conference on Neural Information Processing. LNCS, vol. 6443, pp. 438–445 (2010)Google Scholar
  14. 14.
    Deckert, M.: Batch weighted ensemble for mining data streams with concept drift. In: International Symposium on Methodologies for Intelligent Systems. LNAI, vol. 6804, pp. 290–299 (2011)Google Scholar
  15. 15.
    Yeh, Y., Wang, Y.F.: A rank-one update method for least squares linear discriminant analysis with concept drift. Pattern Recogn. 46(5), 1267–1276 (2013)CrossRefMATHGoogle Scholar
  16. 16.
    Zhu, Q., Hu, X., Zhang, Y., Li, P., Wu, X.: A double-window-based classification algorithm for concept drifting data streams. In: IEEE International Conference on Granular Computing, pp. 639–644 (2010)Google Scholar
  17. 17.
    Alippi, C., Boracchi, G., Roveri, M.: Just-In-Time classifiers for recurrent concepts. IEEE Trans. Neural Netw. 24(4), 620–634 (IEEE) (2013)Google Scholar
  18. 18.
    Li, P., Wu, X., Hu, X.: Mining recurring concept drifts with limited labeled streaming data. ACM Trans. Intell. Syst. Technol. 3(2), 29:1–29:32 (ACM) (2012)Google Scholar
  19. 19.
    Gomes, J.B., Menasalvas, E., Sousa, P.A.: CALDS: Context-aware learning from data streams. In: Proceedings of the First International Workshop on Novel Data Stream Pattern Mining Techniques (StreamKDD’10), pp. 16–24 (2010)Google Scholar
  20. 20.
    Gomes, J.B., Menasalvas, E., Sousa, P.A.: Learning recurring concepts from data streams with a context-aware ensemble. In: Proceedings of the 2011 ACM Symposium on Applied Computing (SAC’11), pp. 994–999 (2011)Google Scholar

Copyright information

© Springer India 2015

Authors and Affiliations

  1. 1.Department of Computer Science and EngineeringPondicherry Engineering CollegePondicherryIndia

Personalised recommendations