Pattern classification with principal component analysis and fuzzy rule bases (Q5928444)

The authors propose a way of reducing the feature space dimension, via the principal components analysis. They show that comparison of the present study with their earlier one [\textit{V. Ravi} and \textit{H.-J. Zimmermann}, ibid. 123, 16-28 (2000; Zbl 0961.90135)] is simply not meaningful because the principal components are only linear combinations of the original feature variables and hence, they do not reflect of the importance or otherwise of the original variables.