Harnessing the Power of Ensemble Algorithms for Diabetes Prediction: A Comparative Analysis

Effective diagnosis of diabetes is crucial for managing the disease and preventing complications. This study explores the use of machine learning for diabetes prediction, focusing on the impact of data preprocessing on model performance. Using four classifiers, namely logistic regression, naive Bayes, random forest, and gradient boosting, diabetes cases are diagnosed after extensive data preprocessing, including cleaning and feature engineering. Results show that the ensemble methods, random forest, and gradient boosting achieve high accuracy and robustness in diabetes prediction, especially with preprocessing. These findings underscore the importance of model selection and preprocessing in healthcare machine learning.