In this paper, we continue our considerations concerning the ranking of intuitionistic fuzzy alternatives (options, variants, ...). We complete our previous considerations by showing in another way why the method proposed by us gives proper results. We stress when the method should be applied and emphasize its transparency.

Dimension reduction of the models, i.e., pointing out only the necessary number of input variables (attributes, features) is an important task enabling the efficient performance of different algorithms. This paper is a continuation of our previous works on a new method of selection of the attributes in the models making use of Atanassov's intuitionistic fuzzy sets. We consider classification problems trying to point out the reduced number of the attributes and still obtain satisfactory results. We investigate the previously proposed method in more details comparing its performance with a well-known method of extraction parameters, namely Principal Component Analysis (PCA), and with a well-known method of selecting the attributes in which the so-called Gain Ratio is used. We illustrate our considerations using benchmark data from UCI Machine Learning Repository.