The paper addresses the problem of scoring and ranking multidimensional ordinal data, by integrating existing “posetic” procedures, with tools from fuzzy logic. In particular, we consider and generalize the recently proposed separation scores, by computing them via triangular norms, the fuzzy logic equivalent of the classical set intersection operator . The resulting scoring procedures is more consistent, from a mathematical point of view, and more flexible, in practical applications. Generalized separation scores are developed and illustrated through the analysis of a small real dataset, pertaining to political pluralism in Eurasia countries.
Fattore, M., De Capitani, L., Avellone, A. (2026). Generalized Separation Scores for Ranking Partially Ordered Data. In F.M. Chelli, C. Crocetta, S. Ingrassia, M.C. Recchioni (a cura di), Statistical Learning, Sustainability and Impact Evaluation SIS 2023, Ancona, Italy, June 21–23 Conference proceedings (pp. 11-24). Springer [10.1007/978-3-032-10630-8_2].
Generalized Separation Scores for Ranking Partially Ordered Data
Fattore, M;De Capitani, L
;Avellone A
2026
Abstract
The paper addresses the problem of scoring and ranking multidimensional ordinal data, by integrating existing “posetic” procedures, with tools from fuzzy logic. In particular, we consider and generalize the recently proposed separation scores, by computing them via triangular norms, the fuzzy logic equivalent of the classical set intersection operator . The resulting scoring procedures is more consistent, from a mathematical point of view, and more flexible, in practical applications. Generalized separation scores are developed and illustrated through the analysis of a small real dataset, pertaining to political pluralism in Eurasia countries.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
