Advances in principal component analysis for intuitionistic fuzzy data sets

We present a novel approach to principal component analysis (PCA) for data expressed in terms of Atanassov's intuitionistic fuzzy sets (A-IFSs), i.e. using the degree of membership, non-membership and hesitation margin which was shown in our works to be a prerequisite for a meaningful analysis of A-IFS type data and information. This new approach to PCA for the A-IFS data is relevant for making possible to better reflect the nature of data and information. Our main focus is the reduction of data dimensionality. An illustrative example on an A-IFS version of the well known Iris data is shown.