Combines fuzzy-c-means feature generation with a neural-network classifier to predict diabetes outcomes on the Pima Indians dataset.
The notebook loads the Pima Indians Diabetes dataset into a DataFrame (768 × 9) via pandas , applies standard scaling, then generates fuzzy membership features using a C-means clustering implementation . It concatenates these fuzzy features with the original scaled inputs to train both a baseline MLP (via TensorFlow/Keras) and a hybrid Neuro-Fuzzy model, comparing their performance through accuracy and other classification metrics.