Logical, algebraic, analytic, and probabilistic aspects of triangular norms
Klement, Erich Peter Mesiar, Radko
Place of Publication
This chapter discusses some specific tools that can be used to build triangular norms based on a finite number of (possibly noisy) observations. Such problem arises in applications, when observed data (e.g., decision patterns of experts) need to be modelled with a special class of functions, such as triangular norms. We show how this problem can be transformed into a constrained regression problem, and then efficiently solved. We also discuss related operators: uninorms, nullnorms and associative copulas.
Field of Research
010107 Mathematical Logic, Set Theory, Lattices and Universal Algebra