Cost-Driven Active Learning with Semi-Supervised Cluster Tree for Text Classification

  • Zhaocai Sun
  • Yunming Ye
  • Yan Li
  • Shengchun Deng
  • Xiaolin Du
Part of the Studies in Computational Intelligence book series (SCI, volume 551)


The key idea of active learning is that it can perform better with less data or costs if a machine learner is allowed to choose the data actively. However, the relation between labeling cost and model performance is seldom studied in the literature. In this paper, we thoroughly study this problem and give a criterion called as cost-performance to balance this relation. Based on the criterion, a cost-driven active SSC algorithm is proposed, which can stop the active process automatically. Empirical results show that our method outperforms active SVM and co-EMT.


Active Learning Semi-supervised Learning Cluster Tree 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Baram, Y., El-Yaniv, R., Luz, K.: Online choice of active learning algorithms. The Journal of Machine Learning Research 5, 255–291 (2004)MathSciNetGoogle Scholar
  2. 2.
    Geman, S., Bienenstock, E., Doursat, R.: Neural networks and the bias/variance dilemma. Neural Computation 4(1), 1–58 (1992)CrossRefGoogle Scholar
  3. 3.
    Lewis, D., Gale, W.A.: A sequential algorithm for training text classifiers. In: ACM SIGIR Conference on Research and Development in Information Retrieval, pp. 3–12 (1994)Google Scholar
  4. 4.
    Nallasamy, U., Metze, F., Schultz, T.: Active learning for accent adaptation in automatic speech recognition. In: IEEE SLT 2012, pp. 360–365 (2012)Google Scholar
  5. 5.
    Niu, B., Cheng, J., Bai, X., Lu, H.: Asymmetric propagation based batch mode active learning for image retrieval. Signal Processing (2012)Google Scholar
  6. 6.
    Patra, S., Bruzzone, L.: A cluster-assumption based batch mode active learning technique. Pattern Recognition Letters 33(9), 1042–1048 (2012)CrossRefGoogle Scholar
  7. 7.
    Quinlan, J.R.: C4. 5: programs for machine learning. Morgan Kaufmann (1993)Google Scholar
  8. 8.
    Roy, N., McCallum, A.: Toward optimal active learning through monte carlo estimation of error reduction. In: ICML 2001 (2001)Google Scholar
  9. 9.
    Schohn, G., Cohn, D.: Less is more: Active learning with support vector machines. In: ICML 2000, pp. 839–846 (2000)Google Scholar
  10. 10.
    Settles, B.: Active learning literature survey. University of Wisconsin, Madison (2010)Google Scholar
  11. 11.
    Settles, B.: Active learning. Synthesis Lectures on Artificial Intelligence and Machine Learning 6(1), 1–114 (2012)CrossRefGoogle Scholar
  12. 12.
    Settles, B., Craven, M., Ray, S.: Multiple-instance active learning. In: Advances in Neural Information Processing Systems, pp. 1289–1296 (2007)Google Scholar
  13. 13.
    Seung, H.S., Opper, M., Sompolinsky, H.: Query by committee. In: Proceedings of the Fifth Annual Workshop on Computational Learning Theory, pp. 287–294 (1992)Google Scholar
  14. 14.
    Steinberg, D., Colla, P.: Cart: classification and regression trees. Salford Systems, San Diego (1997)Google Scholar
  15. 15.
    Su, H., Chen, L., Ye, Y., Sun, Z., Wu, Q.: A refinement approach to handling model misfit in semi-supervised learning. In: Cao, L., Zhong, J., Feng, Y. (eds.) ADMA 2010, Part II. LNCS, vol. 6441, pp. 75–86. Springer, Heidelberg (2010)CrossRefGoogle Scholar
  16. 16.
    Sun, Z., Ye, Y., Zhang, X., Huang, Z., Chen, S., Liu, Z.: Batch-mode active learning with semi-supervised cluster tree for text classification. In: WI 2012, pp. 388–395 (2012)Google Scholar
  17. 17.
    Tong, S., Koller, D.: Support vector machine active learning with applications to text classification. The Journal of Machine Learning Research 2, 45–66 (2002)MATHGoogle Scholar
  18. 18.
    Vapnik, V.N.: The nature of statistical learning theory. Springer (2000)Google Scholar
  19. 19.
    Zhu, J., Ma, M.: Uncertainty-based active learning with instability estimation for text classification. ACM Transactions on Speech and Language Processing 8(4), 5 (2012)CrossRefMathSciNetGoogle Scholar
  20. 20.
    Zhu, X., Lafferty, J., Ghahramani, Z.: Combining active learning and semi-supervised learning using gaussian fields and harmonic functions. In: ICML Workshop on the Continuum from Labeled to Unlabeled Data in Machine Learning and Data Mining, pp. 58–65 (2003)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  • Zhaocai Sun
    • 2
  • Yunming Ye
    • 2
  • Yan Li
    • 1
  • Shengchun Deng
    • 3
  • Xiaolin Du
    • 2
  1. 1.Shenzhen Graduate SchoolHarbin Institute of TechnologyWeihaiChina
  2. 2.School of Computer EngineeringShenzhen PolytechnicShenzhenChina
  3. 3.Department of Computer ScienceHarbin Institute of TechnologyWeihaiChina

Personalised recommendations