In this paper, a new methodology is developed for defining error and similarity measure indexes, in order to establish a criterion adequate for comparison function approximation using fuzzy data. The proposed similarity measures and error indexes are applied for smoothing with cubic splines, showing a good performance for defining the accuracy of approximation obtained with fuzzy numbers. Examples are given to compare the behavior of the new indexes proposed for different configurations of the smooth cubic splines. A statistical analysis was carried out to verify the homogeneity of the indexes proposed as criteria to determine the correctness or accuracy of such approximation of fuzzy numbers. © 2010 Elsevier Inc.
Valenzuela, O., & Pasadas, M. (2011). Fuzzy data approximation using smoothing cubic splines: Similarity and error analysis. Applied Mathematical Modelling, 35(5), 2122–2144. https://doi.org/10.1016/j.apm.2010.11.046