Integration of Machine Learning Algorithms Classifiers and Sequential Forward Selection Features in Diabetes Prediction

Diabetes mellitus, commonly known as diabetes, is a medical condition in which blood sugar levels rise for an extended period. Diabetes is a chronic condition that affects thousands of people. It has resulted in the deaths of millions of people across the world. The only goal of this research is to create a machine-learning model for the early detection of diabetes utilizing supervised machine-learning algorithms and various feature selection strategies for the optimal number of features. On the diabetes dataset obtained from the online platform Kaggle, we employed three distinct machine learning classification algorithms, including Logistic Regression, Decision Tree classifier, and Random Forest classifier. All machine learning models scored roughly 86% accuracy, but the Random Forest model achieved 86.51 % accuracy on a test set with only a subset of four features. The first features are High Blood Pressure, High Cholesterol, Body Mass Index, and General Health, with an accuracy of 86.57% on the train set. The Sequential forward Selection algorithm was applied for the feature.