Independent Component Analysis

From Rapid-I-Wiki

Jump to: navigation, search

Performs an independent component analysis (ICA).



This operator performs the independent componente analysis (ICA). Implementation of the FastICA-algorithm of Hyvaerinen und Oja. The operator outputs a FastICAModel. With the ModelApplier you can transform the features.


  • example set input: expects: ExampleSet


  • example set output:
  • original:
  • preprocessing model:


  • dimensionality reduction:
    Indicates which type of dimensionality reduction should be applied
    Range: none, fixed number; default: none
  • number of components:
    Keep this number of components.
    Range: integer; 1-+?
  • algorithm type:
    If 'parallel' the components are extracted simultaneously, 'deflation' the components are extracted one at a time
    Range: deflation, parallel; default: deflation
  • function:
    The functional form of the G function used in the approximation to neg-entropy
    Range: logcosh, exp; default: logcosh
  • alpha:
    constant in range [1, 2] used in approximation to neg-entropy when fun="logcosh"
    Range: real; 1.0-2.0
  • row norm:
    Indicates whether rows of the data matrix should be standardized beforehand.
    Range: boolean; default: false
  • max iteration:
    maximum number of iterations to perform
    Range: integer; 0-+?; default: 200
  • tolerance:
    A positive scalar giving the tolerance at which the un-mixing matrix is considered to have converged.
    Range: real; 0.0-+?
  • use local random seed:
    Indicates if a local random seed should be used.
    Range: boolean; default: false
  • local random seed:
    Specifies the local random seed
    Range: integer; 1-+?; default: 1992

Personal tools