A Comparative Accuracy Prediction Analysis of Diabetes Disease by Using Machine Learning Algorithms

Abstract: Diabetes is a critical disease that many people suffer from worldwide. It affects the human body's fuel systems. People who are suffering from diabetes have a very high risk of developing diseases like heart disease, kidney disease, stroke, eye problems, nerve damage, etc Generally, eating food is converted into glucose and supplied to different parts of the body through blood. The insulin release by the pancreas helps to balance blood glucose levels. Due to an increased level of glucose in the body, our body stops the production of insulin, which is the leading cause of diabetes. According to the 2022 survey report of t h e Diabetes Federation, 500 million people are presently living with diabetes worldwide. By 2040, this will reach up to 750 million, 01 out of 10 will be living with diabetes. Disease. The research's main objective is to develop an intelligent system that helps in the early prediction of Diabetes and also helps in making correct, accurate, and timely predictive decisions. The key value features are selected by using Pearson's correlation coefficient method. Various machine learning classifiers are used to predict the disease. The random forest algorithm achieved the highest accuracy of 0.83, and Linear regression achieved the lowest accuracy score of 0.64. Keywords: Classification models, supervised machine learning, Random Forest, Linear Regression, SVM