Interval-valued fuzzy regression: Philosophical and methodological issues

This paper revisits interval-valued fuzzy regression and proposes a new unified framework to address type-1 type-2 models. The focuses on two main objectives. First, some philosophical methodological reflections about (IV-T1FR) (IV-T2FR) are discussed analyzed. These aim at positioning avoid misinterpretations that may sometimes lead erroneous or ambiguous considerations in practical applications. Consequently, the interest, relevance, representativeness typology of established. Therefore, IV-T1FR generalizes conventional interval (CIR) increases its specificity. However, if can fit data, then formalism is not able uncertainty phenomenon IV-T1FS representation. In this context, IV-T2FR be regarded as an uncertain IV-T1FR, i.e., generalization environment. Second, methodology models using concepts gradual intervals (GIs) thick (TGIs) proposed. proposed view allows handling via extension standard arithmetic (SIA) – initially for (CIs) GIs TGIs. originality approach resides fact all CIR methodologies extended methods. Furthermore, framework. Our does depend model shape preserves flexibility rigor SIA computations propagation quantities through validated illustrative examples.