Calibrated circular and elliptical regions for intuitionistic fuzzy assessments: A methodological benchmark of geometry coverage and boundary effects
Year: 2026
Type: article
Abstract: Circular and elliptic intuitionistic fuzzy sets motivate region-based descriptions of membership and non-membership uncertainty. This methodological benchmark compares circles, axis-aligned ellipses and oriented ellipses on the admissible intuitionistic triangle. It combines established geometric identities and split-conformal calibration with a reproducible evaluation of coverage, admissible area and parameter variability. Six main scenarios and training-size and coverage-level sensitivity analyses comprise 4,200 independent synthetic datasets and 12,600 fits. The axis-aligned ellipse reduces mean admissible area by 53.4 percent in the main axial-anisotropy scenario; an oriented baseline captures corresponding gains for tilted variation. A full-versus-admissible-area diagnostic isolates the effect of clipping: in the boundary scenario, the oriented ellipse has a 28.5 percent mean-area saving relative to the circle before clipping and a 25.5 percent saving after clipping. Coverage uncertainty accompanies the efficiency comparisons. The contribution is an assessment-domain benchmark and methodological synthesis; it introduces neither a new fuzzy-set algebra nor a new conformal method. Real expert behaviour and downstream decision performance remain outside its evidence.
Authors: Panagiotis Chountas, Georgy Urumov
Institution: University of Westminster
Language: English
Field: Decision Sciences
Domain: Social Sciences
Open Access status: diamond
OpenAlex ID: W7225450299