Machine learning-based early detection of diabetes risk factors for improved health management

This research requires to improve the accuracy of early diabetic forecasting in a human body or patient by applying diverse machine learning approaches. Approaching to creation of machine learning models by using patient datasets to produce predictions with improved accuracy. This work will use machine learning classification and ensemble approaches, such as Random Forest (RF), Gradient Boosting (GB), Decision Tree (DT), K-nearest neighbour (KNN), Logistic Regression (LR), and Support Vector Machine (SVM), on a dataset to predict diabetes. The accuracy of each model differs in comparison to other models. This work demonstrates the model's capability by providing an accurate or greater accuracy. This research paper reported different performance metrics like precision, recall, accuracy, F1 score, and sensitivity for various machine learning algorithms. Final experimental results indicate that the Random Forest classifier outperforms other methods.