Bijective transformations of fuzzy implications -- an algebraic perspective (Q2398048)

In [ibid. 247, 51--67 (2014; Zbl 1334.03028)], the authors of this article studied in the following bijective transformation \(I^\varphi (x,y):=\varphi(I(x,\varphi^{-1}(y)))\), where \(I\) is a fuzzy implication and \(\varphi\) is an increasing bijection on \([0,1]\). They proposed a monoid structure on the set of all fuzzy implications \(\mathbb I\) and, using the largest subgroup \(\mathbb S\) of this monoid, they discussed some group actions on the set \(\mathbb I\). In this article, the authors investigate the following two bijective transformations \(I_\varphi(x,y):=\varphi^{-1}(I(\varphi(x),\varphi(y)))\) and \(\varphi(I)(x,y):=\varphi(I(x,y))\) in a similar way, which were proposed and analysed earlier by \textit{M. Baczyński} and \textit{J. Drewniak} [Lect. Notes Comput. Sci. 1625, 287--298 (1999; Zbl 0929.03034)] and the second author and \textit{R. Mesiar} [Fuzzy Sets Syst. 160, No. 14, 2063--2085 (2009; Zbl 1183.03024)], respectively. The main goal of this paper is to position such bijective transformations from an algebraic perspective. The authors show that these two bijective transformations can also be seen as being obtained from some group actions of \(\mathbb S\) on \(\mathbb I\). Further, they consider the most general bijective transformation that generates fuzzy implications from fuzzy implications of the form \(I_{\varphi,\psi,\mu}(x,y):=\varphi(I(\psi(x),\mu(y)))\) and they show that they can be also obtained as a composition of group actions of \(\mathbb S\) on \(\mathbb I\).