Utilizing Machine Learning for Predicting Diabetes in Pregnant Women: A Comparative Analysis of Logistic Regression, Random Forest, and Naive Bayes Models

Abstract

Diabetes is a significant worldwide health condition that affects millions of people regardless of demographic variations. Diabetes is characterized by increased blood glucose levels that are caused by inadequate insulin synthesis or activity. The purpose of this study is to investigate the use of machine learning and artificial intelligence approaches in the prediction of diabetes, specifically among pregnant women, who are a population that is at a higher risk. Through the use of sophisticated computational methods including Logistic Regression, Random Forest, and Naive Bayes models, the main objective of this work is to forecast the start of diabetes and reduce the difficulties that are connected with it. Out of all of them, the Random Forest method stands out because to its remarkable accuracy rate of 98% on the dataset, which demonstrates its potential for early identification. This research highlights the revolutionary role that machine learning plays in the delivery of prompt medical diagnosis, especially in underprivileged communities where delayed medical treatment often exacerbates health consequences. Through the simplification of the diagnosis procedure, these computational tools make it possible for medical professionals to perform diabetes management in a more effective and proactive manner.