Diabetes Mellitus Prediction using Supervised Machine Learning Techniques

Abstract

Diabetes is a long-term condition that occurs when either the body cannot use insulin properly or the pancreas does not produce sufficient amounts of hormone to control blood glucose levels. High blood sugar levels are a hallmark of diabetes, which belongs to a group of metabolic diseases. The two most prevalent varieties of diabetes are type 1 and type 2, but there are other types as well, such as gestational diabetes, which develops during pregnancy. The number of people with type 1 diabetes has significantly increased. The genetic condition known as type 1 diabetes has a long incubation period and frequently manifests early in life. Cells in people with type 2 diabetes do not properly respond to insulin. It changes over time and mostly depends on how people live their lives. According to a 2022 report by the International Diabetes Federation, currently around 382 million people worldwide have diabetes. By 2035, the Figure is expected to increase to 592 million. One of the most common causes of tissue and organ damage and dysfunction, including blindness, kidney failure, heart failure, and stroke, is diabetes. As a result, early detection of diabetes is critical. This work aims at implementing two machine learning methods like Logistic Regression and Random Forest for diabetes prediction. Each algorithm is calculated to determine the model’s accuracy. Furthermore, the highest accuracy of 99.03% is received by Random Forest.

Ähnliche Arbeiten