B-Matching for Spectral Clustering

  • Tony Jebara
  • Vlad Shchogolev
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4212)


We propose preprocessing spectral clustering with b-matching to remove spurious edges in the adjacency graph prior to clustering. B-matching is a generalization of traditional maximum weight matching and is solvable in polynomial time. Instead of a permutation matrix, it produces a binary matrix with rows and columns summing to a positive integer b. The b-matching procedure prunes graph edges such that the in-degree and out-degree of each node is b, producing a more balanced variant of k-nearest-neighbor. The combinatorial algorithm optimally solves for the maximum weight subgraph and makes subsequent spectral clustering more stable and accurate. Experiments on standard datasets, visualizations, and video data support the use of b-matching to prune graphs prior to spectral clustering.


Video Sequence Weighted Graph Spectral Cluster Kernel Parameter Binary Cluster 
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.
    Chung, F.: Spectral Graph Theory. American Mathematical Society’s CBMS Regional Conference Series in Mathematics, vol. 92 (1997)Google Scholar
  2. 2.
    Edmonds, J.: Paths, trees and flowers. Canadian Journal of Mathematics 17 (1965)Google Scholar
  3. 3.
    Fremuth-Paeger, C., Jungnickel, D.: Balanced network flows. I. A unifying framework for design and analysis of matching algorithms. Networks 33(1), 1–28 (1999)MATHMathSciNetGoogle Scholar
  4. 4.
    Gabow, H.: An efficient reduction technique for degree-constrained subgraph and bidirected network flow problems. In: STOC 1983: Proceedings of the fifteenth annual ACM symposium on theory of computing, pp. 448–456 (1983)Google Scholar
  5. 5.
    Muller-Hannemann, M., Schwartz, A.: Implementing weighted b-matching algorithms: insights from a computational study. J. Exp. Algorithmics 5, 1084–6654 (2000)CrossRefGoogle Scholar
  6. 6.
    Ng., A., Jordan, M., Weiss, Y.: On spectral clustering: Analysis and an algorithm. Neural Information Processing Systems 14 (2001)Google Scholar
  7. 7.
    Papadimitriou, C., Steiglitz, K.: Combinatorial Optimization: Algorithms and Complexity. Prentice-Hall, Englewood Cliffs (1982)MATHGoogle Scholar
  8. 8.
    Roweis, S., Saul, L.: Nonlinear Dimensionality Reduction by Locally Linear Embedding. Science 290, 5500 (2000)CrossRefGoogle Scholar
  9. 9.
    Shi, F., Malik, J.: Normalized Cuts and Image Segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence 22(8), 888–905 (2000)CrossRefGoogle Scholar
  10. 10.
    Tutte, W.T.: A short proof of the factor theorem for finite graphs. Canadian Journal of Mathematics 6, 347–352 (1954)MATHMathSciNetCrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Tony Jebara
    • 1
  • Vlad Shchogolev
    • 1
  1. 1.Department of Computer ScienceColumbia UniversityNew YorkUSA

Personalised recommendations