An Effective Machine Learning Approach for Diabetes Prediction

Abstract

Diabetes is a chronic condition that could lead to a global health care disaster. 382 million people worldwide have diabetes, according to the International Diabetes Federation. This will double to 592 million by 2035. Diabetes is a condition brought on by elevated blood glucose levels. The symptoms of this elevated blood sugar level include frequent urination, increased thirst, and increased hunger. One of the main causes of stroke, kidney failure, heart failure, amputations, blindness, and kidney failure is diabetes. Our bodies convert food into sugars, such as glucose, when we eat. Our pancreas is then expected to release insulin. Insulin acts as a key to unlock our cells, allowing glucose to enter and be used by us as fuel. However, this mechanism does not function in diabetes. The most prevalent forms of the disease are type 1 and type 2, but there are other varieties as well, including gestational diabetes, which develops during pregnancy. Data science has an emerging topic called machine learning that studies how machines learn from experience. The goal of this study is to create a system that, by fusing the findings of several machine learning approaches, can more accurately conduct early diabetes prediction for a patient. K closest neighbour, Logistic Regression, Random Forest, Support Vector Machine, and Decision Tree are some of the techniques employed. Each algorithm's accuracy is calculated along with the model's accuracy. The model for predicting diabetes is then chosen from those with good accuracy.