FID4.1: an overview

FID4.1 is a freeware software tool for supervised classification, implementing fuzzy decision trees (or forest). It is loosely based on the ID3 decision tree algorithm. It handles categorical, continuous, and linguistic attributes and classes. It can also operate with noisy and unknown features, both in training and testing data. For continuous attributes that are not partitioned into fuzzy sets, the system generates fuzzy partitioning using a data-driven top-down or bottom-up approach. Finally, FID4.1 uses a number of inferences, falling into two general classes: set-based and exemplar-based. In this paper we overview the major features and functionalities of FID4.1.