GPU-Accelerated Non-negative Matrix Factorization for Text Mining

  • Volodymyr Kysenko
  • Karl Rupp
  • Oleksandr Marchenko
  • Siegfried Selberherr
  • Anatoly Anisimov
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7337)

Abstract

An implementation of the non-negative matrix factorization algorithm for the purpose of text mining on graphics processing units is presented. Performance gains of more than one order of magnitude are obtained.

Keywords

Text Mining Latent Semantic Analysis Nonnegative Matrix Factorization Vector Space Model Document 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.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Lee, D.D., Seung, H.S.: Algorithms for Non-Negative Matrix Factorization. In: NIPS (2000)Google Scholar
  2. 2.
    Hwu, W.W. (ed.): GPU Computing Gems. MKP (2011)Google Scholar
  3. 3.
  4. 4.
    Khronos Group. OpenCL, http://www.khronos.org/opencl/
  5. 5.
    Agullo, E., et al.: Numerical Linear Algebra on Emerging Architectures: The PLASMA and MAGMA projects. J. Phys.: Conf. Ser. 180(1) (2009)Google Scholar
  6. 6.
  7. 7.
    Rupp, K., et al.: ViennaCL - A High Level Linear Algebra Library for GPUs and Multi-Core CPUs. In: Proc. Intl. Workshop on GPUs and Scientific Applications (GPUScA 2010), pp. 51–56 (2010)Google Scholar
  8. 8.
    Xu, R., Wunsch II, D.: Survey of Clustering Algorithms. IEEE Trans. on Neural Networks 16(3), 645–678 (2005)CrossRefGoogle Scholar
  9. 9.
    Salton, G., et al.: A Vector Space Model for Automatic Indexing. Communications of the ACM 18(11), 613–620 (1975)MATHCrossRefGoogle Scholar
  10. 10.
    Deerwester, S., et al.: Indexing by Latent Semantic Analysis. Journal of the American Society for Information Science 41(6), 391–407 (1990)CrossRefGoogle Scholar
  11. 11.
    Van de Cruys, T.: A non-negative tensor factorization model for selectional preference induction. In: Proc. Workshop on Multiword Expressions, pp. 83–90 (2009)Google Scholar
  12. 12.
    Xu, W., et al.: Document Clustering Based on Non-Negative Matrix Factorization. In: Proc. 26th Intl. Conf. Research and Development in Information Retrieval, pp. 267–273 (2003)Google Scholar
  13. 13.
    Pauca, V., et al.: Text Mining Using Non-Negative Matrix Factorizations. In: Proc. 4th SIAM Intl. Conf. Data Mining (2004)Google Scholar
  14. 14.
    Amy, L., Carl, M.: ALS Algorithms Nonnegative Matrix Factorization Text Mining. SAS NMF Day (2005)Google Scholar
  15. 15.
    Hoyer, P.: Non-Negative Sparse Coding. In: Proc. IEEE Workshop on Neural Networks for Signal Processing (2002)Google Scholar
  16. 16.
    Saad, Y.: Iterative Methods for Sparse Linear Systems. SIAM (2003)Google Scholar
  17. 17.
    An Overview of Eigen. First Plafrim Scientific Day. Bordeaux (May 31, 2011), http://eigen.tuxfamily.org/
  18. 18.

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Volodymyr Kysenko
    • 1
  • Karl Rupp
    • 2
    • 3
  • Oleksandr Marchenko
    • 1
  • Siegfried Selberherr
    • 2
  • Anatoly Anisimov
    • 1
  1. 1.Faculty of CyberneticsTaras Shevchenko National University of KyivUkraine
  2. 2.Institute for MicroelectronicsTU WienAustria
  3. 3.Institute for Analysis and Scientific ComputingTU WienAustria

Personalised recommendations