Diabetes Mellitus Prediction using Machine Learning

Abstract

Abstract: Diabetes is a chronic metabolic disorder affecting millions of people worldwide, and machine learning has shown great potential in predicting the disease using medical and demographic features from patient data. In this paper, we propose a hybrid model of Support Vector Machines (SVM) and XGBoost for diabetes prediction, which combines the strengths of both algorithms to achieve higher accuracy and better performance. We evaluate the proposed model using the Pima Indian diabetes dataset and compare its performance with other machine learning models. To improve the performance of the hybrid model, we also apply feature selection techniques to select the most relevant features from the dataset. Our results show that the proposed hybrid model of SVM and XGBoost achieves higher accuracy, recall, and F1 score compared to other machine learning models such as Logistic Regression, Random Forest, and Naive Bayes. Furthermore, the performance of the hybrid model is significantly improved by feature selection, which helps to reduce the dimensionality of the dataset and focus only on the most relevant features.