FID 3.5: Overview and experimentation

FID is the original fuzzy decision tree, first introduced almost twenty years ago, that sparked a huge variety of hybrid algorithms merging approximate reasoning, fuzzy systems, and mainstream classification algorithms. With the continued interest, this paper describes a newly released update 3.5. One important new addition is a module that can be used to study the effect of noise and missing values on the performance of any classification system - something not well explored in the literature.