Comparative Study of Various Machine Learning Techniques for Diabetes Prediction

Diabetes is a pervasive chronic disease affecting millions worldwide, necessitating effective and early prediction to improve patient outcomes and reduce healthcare costs. This study presents a comparative analysis of various machine learning models in prediction of diabetes, with an emphasis on both individual model performance and ensemble methods. Logistic Regression, Random Forest, Decision Tree, Support Vector Machine (SVM), K-Nearest Neighbors, and XGBoost model’s performance are analyzed using a dataset of clinical health indicators. Accuracy, Precision, Recall, and F1-score are the key metrics considered to determine their predictive model performance. Furthermore, a Voting Classifier is implemented to explore whether an ensemble of models can outperform individual algorithms in terms of predictive accuracy and robustness.Outcomes of the experimental results indicate that the XGBoost model and the Voting Classifier provide superior accuracy and reliability, making them effective tools for early-stage diabetes prediction. This research highlights the potential of machine learning in healthcare and suggests a promising approach to predictive modeling for diabetes diagnosis.