Based on research paper “FSVM-CIL: Fuzzy Support Vector Machines for Class Imbalance Learning” by Rukshan Batuwita and Vasile Palade which discuss Fuzzy concept
It is used for optimazation of algorithm for imbalanced datasets which do not have 1:1 no. of instances of each class.
Pima Indians diabities dataset has 35:65 ratio in their class. haberman, abalone,pageblocks which have diff class ratio.
Preprocessing notebooks are to make a new version of each dataset such that thier class ratio after splitting in test and training is same.