ML - Based Diabetes Foretell Using SVM and Logistic Regression In Healthcare

Abstract

Diabetes is one of the most grievous diseases in the world which has no remedy to cure it after a particular stage. Based on the survey of the last 20 years, the number of people having diabetes tripled. Over 422 million people in the world are diagnosed with diabetes. There are many factors that are responsible for the occurrence of diabetes. It is caused due to increased blood sugar level because of imbalance in insulin processing by the body, which leads to varieties of disorders like Coronary failure, blood pressure, etc and it can also effect other parts of the body. This project mainly focuses on the management of diabetes prediction, that will be approached using machine learning algorithms. Machine learning algorithms provide better results in diabetes detection by constructing models from patient datasets. The aim of this work is to make a prediction of diabetes more precisely with Logistic Regression (binary classification) and Support Vector Machine algorithm(SVM) in machine learning. It predicts the diabetes risk in early stages using symptoms and also predict using distinctive attributes of diabetes. Therefore, two different datasets of patients are used to train the models. This project work will function as an aid for the medical examiners in the diagnosis of diabetes of the patients. Thus, it can significantly help diabetes research and, ultimately, improve the quality of healthcare for diabetic patients.