The Impact of Latency on Online Classification Learning with Concept Drift

  • Gary R. Marrs
  • Ray J. Hickey
  • Michaela M. Black
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6291)


Online classification learners operating under concept drift can be subject to latency in examples arriving at the training base. A discussion of latency and the related notion of example filtering leads to the development of an example life cycle for online learning (OLLC). Latency in a data stream is modelled in a new Example Life-cycle Integrated Simulation Environment (ELISE). In a series of experiments, the online learner algorithm CD3 is evaluated under several drift and latency scenarios. Results show that systems subject to large random latencies can, when drift occurs, suffer substantial deterioration in classification rate with slow recovery.


Online Learning Classification Concept Drift Data stream  Example life-cycle Latency ELISE CD3 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Kolter, J.Z., Maloof, M.A.: Dynamic Weighted Majority: An ensemble method for drifting concepts. Journal of Machine Learning Research 8, 2755–2790 (2007)Google Scholar
  2. 2.
    Minku, L.L., White, A.P., Yao, X.: The Impact of Diversity on On-line Ensemble Learning in the Presence of Concept Drift. IEEE Transactions on Knowledge and Data Engineering, 730–742 (2009)Google Scholar
  3. 3.
    Gao, J., Fan, W., Han, J.: On appropriate assumptions to mine data streams: Analysis and practice. In: Perner, P. (ed.) ICDM 2007. LNCS (LNAI), vol. 4597, pp. 143–152. Springer, Heidelberg (2007)Google Scholar
  4. 4.
    Wang, H., Yin, J., Pei, J., Yu, P., Yu, J.: Suppressing model over-fitting in mining concept-drifting data streams. In: Proc. KDD 2006, Philadelphia, August 20–23, pp. 736–741 (2006)Google Scholar
  5. 5.
    Sculley, D.: Practical learning from one-sided feed-back. In: Proc. of the 13th ACM SIGKDD international conference on Knowledge discovery and data mining, KDD 2007 (2007)Google Scholar
  6. 6.
    Hickey, R.J.: Structure and Majority Classes in Decision Tree Learning. Journal of Machine Learning 8, 1747–1768 (2007)MathSciNetGoogle Scholar
  7. 7.
    Black, M., Hickey, R.J.: Maintaining the performance of a learned classifier under concept drift. Intelligent Data Analysis 3, 453–474 (1999)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2010

Authors and Affiliations

  • Gary R. Marrs
    • 1
  • Ray J. Hickey
    • 1
  • Michaela M. Black
    • 1
  1. 1.School of Computing and EngineeringUniversity of Ulster, ColeraineCounty LondonderryN. Ireland

Personalised recommendations