A Survey based on Early Predicting Diabetes with Machine Learning Algorithms

A chronic medical condition known as diabetes is characterized by high blood glucose levels. Retinopathy, neuropathy, and internal organ failure are just a few of the numerous health effects that diabetes can cause. WHO projections indicate that by 2040, the number could rise to 642 million, or one in ten people, as a result of an unhealthy lifestyle and inactivity. Recently, the medical field has seen a rise in the use of machine learning algorithms to predict the onset of diabetes. Numerous writers have conducted in-depth studies on the use of ML algorithms to predict diabetes. In this paper, we examine the diabetes prediction algorithms developed by multiple authors. PIMA Indian Diabetes Dataset is the foundation for most research. As a result of the investigations, popular machine learning algorithms such as support vector machines (SVM), decision trees, logistic regression, Naive Bayes, random forests, and other classifiers were all examined. It was discovered that the random forest approach outperforms the others, offering an accuracy of 99.5%. Our study is highly advised because it includes articles from a variety of sources that will be useful to future scholars looking at different diabetic prediction models.