An Ensemble Learning Approach for Effective Prediction of Diabetes Mellitus Using Hard Voting Classifier

Objectives: People all across the world are afflicted by deadly ailment known as diabetes. Diabetes is a terrible condition characterized high blood glucose levels. This chronic one of leading causes death for people worldwide. Early identification and prediction diabetes can be aided machine learning techniques. The purpose this study to use an ensemble algorithms predict efficiently in order help patients suffering from lethal disease. Methods: existing methods single model diabetes, which may have impact on accuracy because no fit datasets. Therefore we propose robust based using hard voting classifier. Both Pima Indians dataset Stage Risk Prediction Dataset, collect data with without were tested. For classification, proposed classifier uses combination three namely logistic regression, decision tree, support vector machine. Findings: On PIMA dataset, approach achieves highest accuracy, precision, recall, F1 score value 81.17%, while it 94.23%. Novelty: methodology was experimentally tested state-of-the-art technology basic classifiers such K-Nearest Neighbor, Logistic Regression, Support Vector Machine, Random Forest. results validated computing confusion matrix ROC each classier type. Keywords: Detection; Machine Learning; Supervised Classification; Ensemble Hard Voting Classifier