Entropy-Based Symmetric Implicational Method for R-and (S,N)-Implications

Yiming Tang,Xiaocheng Song,Fuji Ren,Witold Pedrycz
DOI: https://doi.org/10.1109/ISKE47853.2019.9170284
2019-01-01
Abstract:The criterion of largest entropy can provide a suitable interpretation for how to choose the optimal outcome of reasoning mechanism. Enlightened by such idea, in this study the entropy-based symmetric implicational (EBSI) method is presented. First of all, the novel EBSI criterion is provided, which is embodied as a kind of enhanced form of the fuzzy entropy full implication criterions. Moreover, the fundamental properties of outcome of such strategy are explored. Furthermore, unified outcomes of such strategy are constructed, which is based upon some properties of these fuzzy implications. Then the ideal outcomes are given for some kinds of implications and particular fuzzy implications.
What problem does this paper attempt to address?