Akademik Veri Yönetim Sistemi
21st International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems (IPMU 2026), Rome, İtalya, 15 - 19 Haziran 2026, ss.177-182, (Tam Metin Bildiri)
This paper examines similarity-based reasoning systems from the perspective of similarity relations and extensionality. Using the concepts of extensional hulls and extensional bases, the behavior of particular matching functions combined with appropriately chosen modi cation functions and aggregation functions may lead not only to robustness, but also to the preservation of the fuzzy interpolation condition. And even more, as, due to the robustness, the outputs equivalent to particular consequents are obtained not only for the inputs equal to the corresponding antecedents, but also for a broader class of fuzzy sets. Namely, to the fuzzy sets closed under the extensional hull or extensional base operations. The results extend classical fuzzy interpolation to the similarity-based reasoning framework and provide guidance for designing interpretable and reliable inference systems.