Integrating Novel Class Detection with Classification for Concept-Drifting Data Streams

  • Mohammad M. Masud
  • Jing Gao
  • Latifur Khan
  • Jiawei Han
  • Bhavani Thuraisingham
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5782)


In a typical data stream classification task, it is assumed that the total number of classes are fixed. This assumption may not be valid in a real streaming environment, where new classes may evolve. Traditional data stream classification techniques are not capable of recognizing novel class instances until the appearance of the novel class is manually identified, and labeled instances of that class are presented to the learning algorithm for training. The problem becomes more challenging in the presence of concept-drift, when the underlying data distribution changes over time. We propose a novel and efficient technique that can automatically detect the emergence of a novel class in the presence of concept-drift by quantifying cohesion among unlabeled test instances, and separation of the test instances from training instances. Our approach is non-parametric, meaning, it does not assume any underlying distributions of data. Comparison with the state-of-the-art stream classification techniques prove the superiority of our approach.


Data Stream Leaf Node Test Instance Decision Boundary Ensemble Size 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


  1. 1.
    Chen, S., Wang, H., Zhou, S., Yu, P.: Stop chasing trends: Discovering high order models in evolving data. In: Proc. ICDE, pp. 923–932 (2008)Google Scholar
  2. 2.
    Hulten, G., Spencer, L., Domingos, P.: Mining time-changing data streams. In: Proc. ACM SIGKDD, pp. 97–106 (2001)Google Scholar
  3. 3.
    Kolter, J., Maloof, M.: Using additive expert ensembles to cope with concept drift. In: Proc. ICML, pp. 449–456 (2005)Google Scholar
  4. 4.
    Markou, M., Singh, S.: Novelty detection: A review-part 1: Statistical approaches, part 2: Neural network based approaches. Signal Processing 83 (2003)Google Scholar
  5. 5.
    Masud, M.M., Gao, J., Khan, L., Han, J., Thuraisingham, B.: A practical approach to classify evolving data streams: Training with limited amount of labeled data. In: Proc. ICDM, pp. 929–934 (2008)Google Scholar
  6. 6.
    Roberts, S.J.: Extreme value statistics for novelty detection in biomedical signal processing. In: Proc. Int. Conf. on Advances in Medical Signal and Information Processing, pp. 166–172 (2000)Google Scholar
  7. 7.
    Spinosa, E.J., de Leon, A.P., de Carvalho, F., Gama, J.: Olindda: a cluster-based approach for detecting novelty and concept drift in data streams. In: Proc. 2007 ACM symposium on Applied computing, pp. 448–452 (2007)Google Scholar
  8. 8.
    University of Texas at Dallas Technical report UTDCS-13-09 (June 2009),
  9. 9.
    Wang, H., Fan, W., Yu, P., Han, J.: Mining concept-drifting data streams using ensemble classifiers. In: Proc. ACM SIGKDD, pp. 226–235 (2003)Google Scholar
  10. 10.
    yan Yeung, D., Chow, C.: Parzen-window network intrusion detectors. In: Proc. International Conference on Pattern Recognition, pp. 385–388 (2002)Google Scholar
  11. 11.
    Yang, Y., Wu, X., Zhu, X.: Combining proactive and reactive predictions for data streams. In: Proc. ACM SIGKDD, pp. 710–715 (2005)Google Scholar
  12. 12.
    Yang, Y., Zhang, J., Carbonell, J., Jin, C.: Topic-conditioned novelty detection. In: Proc. ACM SIGKDD, pp. 688–693 (2002)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2009

Authors and Affiliations

  • Mohammad M. Masud
    • 1
  • Jing Gao
    • 2
  • Latifur Khan
    • 1
  • Jiawei Han
    • 2
  • Bhavani Thuraisingham
    • 1
  1. 1.University of TexasDallas
  2. 2.University of IllinoisUrbana Champaign

Personalised recommendations