Elkan’s k-Means Algorithm for Graphs

  • Brijnesh J. Jain
  • Klaus Obermayer
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6438)


This paper proposes a fast k-means algorithm for graphs based on Elkan’s k-means for vectors. To accelerate the k-means algorithm for graphs without trading computational time against solution quality, we avoid unnecessary graph distance calculations by exploiting the triangle inequality of the underlying distance metric. In experiments we show that the accelerated k-means for graphs is faster than k-means for graphs provided there is a cluster structure in the data.


clustering k-means graph matching 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Jiang, X., Munger, A., Bunke, H.: On Median Graphs:Properties, Algorithms, and Applications. IEEE Trans. PAMI 23(10), 1144–1151 (2001)Google Scholar
  2. 2.
    Dosch, P., Valveny, E.: Report on the second symbol recognition contest. In: Liu, W., Lladós, J. (eds.) GREC 2005. LNCS, vol. 3926, pp. 381–397. Springer, Heidelberg (2006)CrossRefGoogle Scholar
  3. 3.
    Elkan, C.: Using the triangle inequality to accelerate k-means. In: ICML 2003 Conference Proceedings, pp. 147–153 (2003)Google Scholar
  4. 4.
    Ferrer, M.: Theory and algorithms on the median graph. application to graph-based classification and clustering, PhD Thesis, Univ. Autònoma de Barcelona (2007)Google Scholar
  5. 5.
    Gold, S., Rangarajan, A.: Graduated Assignment Algorithm for Graph Matching. IEEE Trans. PAMI 18, 377–388 (1996)Google Scholar
  6. 6.
    Hochbaum, D., Shmoys, D.: A best possible heuristic for the k-center problem. Mathematics of Operations Research 10(2), 180–184 (1985)MathSciNetCrossRefzbMATHGoogle Scholar
  7. 7.
    Indyk, P., Amir, A., Efrat, A., Samet, H.: Efficient algorithms and regular data structures for dilation, location and proximity problems. In: FOCS 1999 Conference Proceedings, pp. 160–170 (1999)Google Scholar
  8. 8.
    Jain, B., Wysotzki, F.: Central Clustering of Attributed Graphs. Machine Learning 56, 169–207 (2004)CrossRefzbMATHGoogle Scholar
  9. 9.
    Jain, B., Obermayer, K.: On the sample mean of graphs. In: IJCNN 2008 Conference Proceedings, pp. 993–1000 (2008)Google Scholar
  10. 10.
    Jain, B., Obermayer, K.: Algorithms for the sample mean of graphs. In: Jiang, X., Petkov, N. (eds.) CAIP 2002. LNCS, vol. 5702, pp. 351–359. Springer, Heidelberg (2009)CrossRefGoogle Scholar
  11. 11.
    Jain, B., Obermayer, K.: Structure Spaces. Journal of Machine Learning Research 10 (2009)Google Scholar
  12. 12.
    Jain, B., Obermayer, K.: Graph Quantization. arXiv:1001.0921v[1cs.AI] (2010)Google Scholar
  13. 13.
    Kazius, J., McGuire, R., Bursi, R.: Derivation and validation of toxicophores for mutagenicity prediction. Journal of Medicinal Chemistry 48(1), 312–320 (2005)CrossRefGoogle Scholar
  14. 14.
    Moore, A.: The anchors hierarchy: Using the triangle inequality to survive high dimensional data. In: UAI 2000 Conference Proceedings, pp. 397–405 (2000)Google Scholar
  15. 15.
    Riesen, K., Bunke, H.: IAM Graph Database Repository for Graph Based Pattern Recognition and Machine Learning. In: SSPR 2008 Conference Proceedings, pp. 287–297 (2008)Google Scholar
  16. 16.
    Schenker, A., Last, M., Bunke, H., Kandel, A.: Clustering of web documents using a graph model. In: Web Document Analysis: Challenges and Opportunities, pp. 1–16 (2003)Google Scholar
  17. 17.
    Theodoridis, S., Koutroumbas, K.: Pattern Recognition. Elsevier, Amsterdam (2009)zbMATHGoogle Scholar
  18. 18.
    Watson, C., Wilson, C.: NIST Special Database 4, Fingerprint Database, National Institute of Standards and Technology (1992)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2010

Authors and Affiliations

  • Brijnesh J. Jain
    • 1
  • Klaus Obermayer
    • 1
  1. 1.Berlin Institute of TechnologyGermany

Personalised recommendations