A Graph Based Approach to Multiview Clustering

  • Moumita Saha
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8251)


Rich and complex data sets prevalent in many applications can often be explored from multiple perspectives. Examples include clustering of multimedia, multilingual and heterogeneously linked data sets. Multiview clustering attempts to discover clusters from different views of the same data set. In this article, construction of subspace representation of the views and subsequently clustering the subspaces to produce multiview clusters is been proposed. The subspaces are obtained by a separate clustering procedure on the nearest neighbour graphs of the individual features. Three graph similarity measures are used for this clustering. Empirical results on three benchmark data sets shows that the proposed method provides superior performance in terms of classification accuracy using known class labels as compared to single view clustering of the entire data sets.


Multiview clustering graph measures subspace clustering 


  1. 1.
    Bickel, S., Scheffer, T.: Multi-View Clustering. In: Proceeding of 4th IEEE International Conference on Data Mining, pp. 19–26 (2004)Google Scholar
  2. 2.
    Chi, M., Zhang, P., Zhao, Y., Feng, R., Xue, X.: Web Image Re-Ranking with Multi-view Clustering. In: Proceeding of 18th Int. World Wide Web Conference, pp. 1189–1190. ACM (2009)Google Scholar
  3. 3.
    Kim, Y.-M., Reza Amini, M., Goutte, C., Gallinari, P.: Multi-View Clustering of Multilingual Documents. In: Proceeding of 33rd International ACM SIGIR Conference on Research and Development in Information Retrieval, pp. 821–822 (2010)Google Scholar
  4. 4.
    Greene, D., Cunningham, P.: Multi-View Clustering for Mining Heterogeneous Social Network Data. In: Proceeding of Workshop on Information Retrieval over Social Networks, 31st European Conference on Information Retrieval, ECIR 2009 (2009)Google Scholar
  5. 5.
    Bruno, E., Maillet, S.M.: Multiview Clustering: A Late Fusion Approach using Latent Models. Proceedings of 32nd International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2009, 736–737 (2009)Google Scholar
  6. 6.
    Chaudhuri, K., Kakade, S.M., Livescu, K., Sridharan, K.: Multi-View Clustering via Canonical Correlation Analysis. In: Proceedings of 26th Annual International Conference on Machine Learning, ICML 2009, pp. 129–136 (2009)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • Moumita Saha
    • 1
  1. 1.Department of Computer Science and EngineeringIndian Institute of TechnologyKharagpurIndia

Personalised recommendations