Modeling the Uncertainty of a Set of Graphs Using Higher-Order Fuzzy Sets

摘要: Recent advances in type-2 fuzzy sets (T2FS) have attracted considerable attention for applications data mining and pattern recognition. In particular, there is an effort designing granulation procedures able to generate, from raw input measurements, data, granules of information modeled as T2FS. From our viewpoint, the principal aim those embed into generated T2FS model key uncertainty characterizing data. However, date no formal principle or guideline evaluation such these terms. this paper, define a framework design evaluate what we called uncertainty-preserving transformation procedures, which are basically computational that chapter, deal with measurements represented graphs; hence, set graphs \(\mathcal{G}\) seen sampled unknown generating process P. The is, however, meant be general thus applicable any type. We motivate explain proposed by performing experimental evaluations on ad hoc synthetically datasets.