Implementation of Web Based Intelligent System for Diabetes Prediction Using Machine Learning

The study focuses on the development and implementation of a web based intelligent system which predicts diabetes with improved accuracy matrices than previous studies. The process of prediction of heart disease involves 6 different models of machine learning SVM, KNN, Naïve Bayes, Logistic Regression, Decision Tree and XGBoost. The early diagnosis and intervention to prevent severe complications is necessary. The study contributes to the society in the field of predictive healthcare by demonstrating the feasibility and effectiveness of such smart disease predictions models in a web-based platform. The model is capable of generating customized reports with exceptional memory, precision, and confusion matrix by allowing patients to input their symptoms through the application's interface. The results of the study showed that XGBoost model really outperformed in terms of accuracy with 89% while Naïve Bayes model does not performed much effectively on our dataset as accuracy of this model is very less as compared to other models that we have implemented on our dataset for the prediction of diabetes.