Advertisement

A Novel Approach to Overcome the Limitations of Power Iteration Algorithm Designed for Clustering

  • D. JayalatchumyEmail author
  • P. Thambidurai
  • D. Kadhirvelu
Conference paper
Part of the Lecture Notes in Networks and Systems book series (LNNS, volume 98)

Abstract

The rise of Big Data has generated a new challenge to develop and refine algorithms to process and extract useful information. Power Iteration Clustering is the most efficient and scalable algorithms that can deal with larger datasets. It uses the power method for finding the eigen values and vectors. Despite its advantages the main disadvantage is that the convergence of this method relies on the magnitude of the leading eigen values that do not guarantee convergence. Moreover, it gives only one pseudo eigen value and it seems to be difficult when it is multideimensional. Hence, the algorithm has been refined using various acceleration techniques to make it more efficient to handle larger datasets. Various experiments was conducted to check its validity and it has been proved that the power method modified is efficient and suitable for processing larger datasets more accurately.

Keywords

PIC Aitken Steffensen Banach fixed point Cauchy theorem 

References

  1. 1.
    Lin, F., Cohen, W.: Power iteration clustering. In: Proceedings of the 27th International Conference on Machine Learning, Haifa (2010)Google Scholar
  2. 2.
    Jia, H., Ding, H., Xu, X., Nie, R.: The latest research progress on Spectral Clustering. J. Neural Comput. Appl. 24(7–8), 1477–1486 (2014)CrossRefGoogle Scholar
  3. 3.
    Kamvar, S.D., Golub, G.H., et al.: Extrapolation methods for accelerating PageRank computations. In: Proceedings of the 12th International Conference on the World Wide Web, pp. 261–270. ACM (2003)Google Scholar
  4. 4.
    Yan, W., et al.: p-PIC: parallel power iteration clustering for big data. J. Parallel Distrib. Comput. Models Algorithms High Perform. Distrib. Data Min. 73(3), 352–359 (2013)Google Scholar
  5. 5.
    Eyi, S.: Convergence error estimation and convergence acceleration in iteratively. In: Proceedings, Eyi Convergence EE (2012)Google Scholar
  6. 6.
    Collomb, C.: A tutorial on the Aitken convergence acceleratorGoogle Scholar
  7. 7.
  8. 8.
  9. 9.
    HELM (Helping Engineers Learn Mathematics) Workbooks: Eigenvalues and eigenvectors, Helping Engineers Learn Mathematics, Loughborough (2008)Google Scholar
  10. 10.
  11. 11.
    Lambers, J.: Fixed Point Iteration. Fall (2009). http://www.math.usm.edu/lambers/mat460/fall09/lecture9.pdf
  12. 12.
    Rousseau, C.: Banach fixed point theorem and applications (2010). http://dmuw.zum.de/images/b/bd/Banach2.pdf
  13. 13.
    Lambers, J.: Accelerating Convergence. Numerical Analysis I. Lecture 13 notes MAT460/560 (2009). http://www.math.usm.edu/lambers/mat460/fall09/lecture13.pdf
  14. 14.
    Cordero, A., Hueso, J.L., et al.: Steffensen type methods for solving nonlinear equations. J. Comput. Appl. Math. 236, 3058–3064 (2012)MathSciNetCrossRefGoogle Scholar
  15. 15.
  16. 16.

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  • D. Jayalatchumy
    • 1
    Email author
  • P. Thambidurai
    • 1
  • D. Kadhirvelu
    • 2
  1. 1.Perunthalaivar Kamarajar Institute of Engineering and TechnologyKaraikalIndia
  2. 2.Sunshine InstitutionPondicherryIndia

Personalised recommendations