Advertisement

Instance Selection Using Two Phase Collaborative Neighbor Representation

  • Fadi Dornaika
  • I. Kamal Aldine
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8681)

Abstract

Finding relevant instances in databases has always been a challenging task. Recently a new method, called Sparse Modeling Representative Selection (SMRS) has been proposed in this area and is based on data self-representation. SMRS estimates a matrix of coefficients by minimizing a reconstruction error and a regularization term on these coefficients using the L 1,q matrix norm. In this paper, we propose another alternative of coding based on a two stage Collaborative Neighbor Representation in which a non-dense matrix of coefficients is estimated without invoking any explicit sparse coding. Experiments are conducted on summarizing a video movie and on summarizing training face datasets used for face recognition. These experiments showed that the proposed method can outperform the state-of-the art methods.

Keywords

Instance selection collaborative neighbor representation video summarization classification 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Garcia, S., Derrac, J., Cano, R., Herrera, F.: Prototype selection for nearest neighbor classification: Taxonomy and empirical study. IEEE Transactions on Pattern Analysis and Machine Intelligence 34(3), 417–435 (2012)CrossRefGoogle Scholar
  2. 2.
    Gu, M., Eisenstat, S.: Efficient algorithms for computing a strong rankrevealing qr factorization. SIAM Journal on Scientific Computing 17, 848–869 (1996)CrossRefzbMATHMathSciNetGoogle Scholar
  3. 3.
    Frey, B., Dueck, D.: Clustering by passing messages between data points. Science Magazine 315, 972–976 (2007)zbMATHMathSciNetGoogle Scholar
  4. 4.
    Tropp, J.: Column subset selection, matrix factorization and eigenvalue optimization. In: Proc. of ACM-SIAM Symposium on Discrete Algorithms (SODA), pp. 978–986 (January 2009)Google Scholar
  5. 5.
    Boutsidis, C., Mahoney, M., Drineas, P.: An improved approximation algorithm for the column subset selection problem. In: Proc. of ACM-SIAM Symposium on Discrete Algorithms (SODA), pp. 968–977 (January 2009)Google Scholar
  6. 6.
    Elhamifar, E., Sapiro, G., Vidal, R.: See all by looking at a few: Sparse modeling for finding representative objects. In: Proc. of IEEE Conference on Computer Vision and Pattern Recognition, pp. 1600–1607 (June 2012)Google Scholar
  7. 7.
    Bien, J., Xu, Y., Mahoney, M.: CUR from a sparse optimization viewpoint. In: Advances in Neural Information Processing Systems, pp. 217–225 (December 2010)Google Scholar
  8. 8.
    Chan, T.: Rank revealing qr factorizations. Linear Algebra and its Applications 88–89, 67–82 (1987)Google Scholar
  9. 9.
    Esser, E., Moller, M., Osher, S., Sapiro, G., Xin, J.: A convex model for nonnegative matrix factorization and dimensionality reduction on physical space. IEEE Transactions on Image Processing 21(7), 3239–3252 (2012)CrossRefMathSciNetGoogle Scholar
  10. 10.
    Charikar, M., Guha, S., Tardos, A., Shmoys, D.: A constant-factor approximation algorithm for the k-median problem. Journal of Computer System Sciences 65(1), 129–149 (2002)CrossRefzbMATHMathSciNetGoogle Scholar
  11. 11.
    Givoni, I., Chung, C., Frey, B.: Hierarchical affinity propagation. In: Conference on Uncertainty in Artificial Intelligence (July 2011)Google Scholar
  12. 12.
    Duda, R., Hart, P., Stork, D.: Pattern Classification. Wiley-Interscience (2004)Google Scholar
  13. 13.
    Dueck, D., Frey, B.: Non-metric affinity propagation for unsupervised image categorization. In: Proc. of International Conference in Computer Vision, pp. 1–8 (October 2007)Google Scholar
  14. 14.
    Waqas, J., Yi, Z., Zhang, L.: Collaborative neighbor representation based classification using l 2-minimization approach. Pattern Recognition Letters 34(2), 201–208 (2013)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  • Fadi Dornaika
    • 1
    • 2
  • I. Kamal Aldine
    • 1
  1. 1.University of the Basque Country EHU/UPVSan SebastianSpain
  2. 2.IKERBASQUE, Basque Foundation for ScienceBilbaoSpain

Personalised recommendations