A Hybrid Meta-Classifier of Fuzzy Clustering and Logistic Regression for Diabetes Prediction

Diabetes is a chronic health condition that impairs the body's ability to convert food energy, recognized by persistently high levels of blood glucose. Undiagnosed diabetes can cause many complications, including retinopathy, nephropathy, neuropathy, and other vascular disorders. Machine learning methods be very useful for disease identification, prediction, treatment. This paper proposes new ensemble approach type 2 prediction based on hybrid meta-classifier fuzzy clustering logistic regression. The proposed consists two levels. First, base-learner comprising six machine algorithms utilized predicting diabetes. Second, meta-learner combines regression employed appropriately integrate predictions from base-learners provide an accurate employs Fuzzy C-means Clustering (FCM) algorithm generate highly significant clusters base-learners. their are then as inputs Logistic Regression (LR) algorithm, which generates final result. Experiments were conducted using publicly available datasets, Pima Indians Database (PIDD) Schorling Dataset (SDD) demonstrate efficacy method When compared with models, outperformed them obtained highest accuracies 99.00% 95.20% PIDD SDD respectively.