Prediction of Diabetes Disease during Pregnancy Using Machine Learning

Abstract

In recent years, diseases associated with diabetes have emerged as one of the leading causes of death globally. In today's world, diabetes accounts for approximately one death every minute. This research mainly aims to determine which patients are at higher risk of developing diabetes based on different medical traits. A significant goal of this study is to create a system for disease awareness and prediction. Recently, various researchers have employed machine learning methods to aid healthcare professionals and specialists in diagnosing diabetes. This research highlights the use of supervised learning methods, such as Linear Regression, Naive Bayes, Decision Trees, and Logistic Regression, with the proposed model reaching an accuracy rate of 86%. The research concentrates on five significant parameters associated with diabetes: blood sugar level, blood pressure, cholesterol, uric acid, and hypertension. These factors are evaluated to understand their influence on the onset of diabetes. The findings illustrate successful outcomes using these algorithms. While diabetes is not curable, it can be effectively managed if detected early. Machine Learning (ML) facilitates the automated early identification of diabetes and has shown to be more efficient and precise compared to conventional human diagnosis.