An Evaluation on Different Graphs for Semi-supervised Learning

  • Chun-guang Li
  • Xianbiao Qi
  • Jun Guo
  • Bo Xiao
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7202)


Graph-based Semi-Supervised Learning (SSL) has been an active topic in machine learning for about a decade. It is well-known that how to construct the graph is the central concern in recent work since an efficient graph structure can significantly boost the final performance. In this paper, we present a review on several different graphs for graph-based SSL at first. And then, we conduct a series of experiments on benchmark data sets in order to give a comprehensive evaluation on the advantageous and shortcomings for each of them. Experimental results shown that: a) when data lie on independent subspaces and the number of labeled data is enough, the low-rank representation based method performs best, and b) in the majority cases, the local sparse representation based method performs best, especially when the number of labeled data is few.


Heat Kernel Sparse Representation Label Data Unlabeled Data Label Propagation 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Szummer, M., Jaakkola, T.: Partially labeled classification with markov random walks, vol. 14. MIT Press (2001)Google Scholar
  2. 2.
    Belkin, M., Niyogi, P., Sindhwani, V.: Manifold regularization: A geometric framework for learning from labeled and unlabeled examples. Journal of Machine Learning Research 7, 2399–2434 (2006)MathSciNetzbMATHGoogle Scholar
  3. 3.
    Zhu, X., Ghahramani, Z., Lafferty, J.D.: Semi-supervised learning using gaussian fields and harmonic functions. In: ICML, pp. 912–919 (2003)Google Scholar
  4. 4.
    Zhou, D., Bousquet, O., Lal, T.N., Weston, J., Schölkopf, B.: Learning with local and global consistency. In: Advances in Neural Information Processing Systems, vol. 16, pp. 321–328. MIT Press (2004)Google Scholar
  5. 5.
    Bengio, Y., Delalleau, O., Le Roux, N.: Label propagation and quadratic criterion. In: Chapelle, O., Schölkopf, B., Zien, A. (eds.) Semi-Supervised Learning, pp. 193–216. MIT Press (2006)Google Scholar
  6. 6.
    Zhou, D., Schölkopf, B., Hofmann, T.: Semisupervised learning on directed graphs. In: NIPS (2005)Google Scholar
  7. 7.
    Wang, F., Zhang, C.: Label propagation through linear neighborhoods. IEEE Transactions on Knowledge and Data Engineering 20(1), 55–67 (2008)CrossRefGoogle Scholar
  8. 8.
    Yan, S., Wang, H.: Semi-supervised learning by sparse representation. In: SIAM International Conference on Data Mining (2009)Google Scholar
  9. 9.
    Cheng, H., Liu, Z., Yang, J.: Sparsity induced similarity measure for label propagation. In: ICCV (2009)Google Scholar
  10. 10.
    Belkin, M., Niyogi, P.: Laplacian eigenmaps for dimensionality reduction and data representation. Neural Computation 15(6), 1373–1396 (2003)zbMATHCrossRefGoogle Scholar
  11. 11.
    Roweis, S.T., Saul, L.K.: Nonlinear dimensionality reduction by locally linear embedding. Science 290(5500), 2323–2326 (2000)CrossRefGoogle Scholar
  12. 12.
    Saul, L.K., Roweis, S.T.: Think globally, fit locally: unsupervised learning of low dimensional manifolds. Journal of Machine Learning Research 4, 119–155 (2003)MathSciNetGoogle Scholar
  13. 13.
    Wang, F., Zhang, C., Shen, H., Wang, J.: Semi-supervised classification using linear neighborhood propagation. In: Proc. IEEE Computer Society Conference on Computer Vision and Pattern Recognition, vol. 1, pp. 160–167 (2006)Google Scholar
  14. 14.
    Li, C.G., Guo, J. Zhang, H.G.: Local sparse representation based classification. In: ICPR (2010)Google Scholar
  15. 15.
    Liu, G., Lin, Z., Yu, Y.: Robust subspace segmentation by low-rank representation. In: ICML (2010)Google Scholar
  16. 16.
    Elhamifar, E., Vidal, R.: Sparse subspce clustering. In: CVPR (2009)Google Scholar
  17. 17.
    Georghiades, A., Belhumeur, P., Kriegman, D.: From few to many: Illumination cone models for face recognition under variable lighting and pose. IEEE Trans. Pattern Anal. Mach. Intelligence 23(6), 643–660 (2001)CrossRefGoogle Scholar
  18. 18.
    Lee, K., Ho, J., Kriegman, D.: Acquiring linear subspaces for face recognition under variable lighting. IEEE Trans. Pattern Anal. Mach. Intelligence 27(5), 684–698 (2005)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Chun-guang Li
    • 1
  • Xianbiao Qi
    • 1
  • Jun Guo
    • 1
  • Bo Xiao
    • 1
  1. 1.PRIS Lab.Beijing University of Posts and TelecommunicationsBeijingChina

Personalised recommendations