Abstract
This paper proposes an enhanced technique of independent component analysis (ICA), which extracts independent components uncorrelated to some external criteria. Fast ICA algorithm is performed after the preprocessing by principal component analysis with external criteria, in which the effects of the external criteria are removed from data sets using regression analysis. The proposed method can also be expanded into local ICA by using hybrid technique of fuzzy clustering and regression analysis. Numerical experiments including knowledge discovery from POS transaction data reveal the characteristic feature of the proposed method.
