• DocumentCode
    698333
  • Title

    Effect of normalization of eigenvectors on the past and RP algorithms for PCA

  • Author

    Landqvist, R. ; Mohammed, A.

  • Author_Institution
    Sch. of Eng., Blekinge Inst. of Technol., Ronneby, Sweden
  • fYear
    2005
  • fDate
    4-8 Sept. 2005
  • Firstpage
    1
  • Lastpage
    4
  • Abstract
    This paper investigates the effect of incorporating normalization of the eigenvectors between iterations in the PAST and RP algorithms for principal component analysis (PCA). In addition, an algorithm denoted as exact eigendecomposition (EE) is proposed for PCA. The algorithms are compared for different configurations using Monte Carlo simulations. Simulation results show that EE has the best performance and that normalization may be used for improving PAST and RP.
  • Keywords
    Monte Carlo methods; eigenvalues and eigenfunctions; principal component analysis; signal processing; Monte Carlo simulations; PCA; eigenvectors; exact eigendecomposition; normalization; principal component analysis; Algorithm design and analysis; Correlation; Eigenvalues and eigenfunctions; Monte Carlo methods; Principal component analysis; Signal processing algorithms; Vectors;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Signal Processing Conference, 2005 13th European
  • Conference_Location
    Antalya
  • Print_ISBN
    978-160-4238-21-1
  • Type

    conf

  • Filename
    7077916