Web Application-based Diabetes Prediction using Machine Learning

Diabetes is a crucial illness, and a lot of people are struggling with it. People with diabetic issues are at high risk of illnesses like cardiovascular disease, kidney illness, stroke, eye issues, tissue damage, and so on. The present exercise in a medical facility is to gather needed information for diabetic issues through medical diagnosis with different examinations, and suitable therapy is offered based upon the medical diagnosis, which takes more effort and resources. However, by using machine learning, there is a solution to this serious problem. In the proposed work, the machine learning algorithms K-Nearest Neighbor (KNN) and Random Forest (RF) were used to predict the risk of developing diabetes-related diseases. After pre-processing the data, features are selected based on their influence on the prediction of the diseases. The prediction accuracy using Random Forest (RF) on the preprocessed data after feature selection and applying the ensemble classifier is 83%, which is significantly higher than the K-Nearest Neighbor (KNN) with an accuracy of 75%. Then, these machine learning models were used to create a web application for diabetes prediction. The objective is to use machine learning techniques to predict the risk of developing diabetes-related diseases and create a web application to support this prediction.