Prediction of Diabetes Using Various Machine Learning Techniques

Abstract

Diabetes is a life-threatening disease marked by unusually high blood sugar levels. It is the leading cause of death in the globe. According to rising morbidity in recent years, the number of diabetic patients globally will reach 642 million by 2040, or approximately one out of every ten persons. It is true that this requires a lot of focus. On the diabetes dataset, a number of data mining and machine learning techniques were utilized to predict disease risk. The goal of this work is to investigate several machine learning algorithms for diabetes categorization, early-stage identification, and prediction using a feature-based dataset. A benchmark PIMA Indian Diabetes dataset is used for experimental evaluation, which includes 768 patients, 268 of whom are diabetic and 500 of whom are not. At the end, the accuracy of various machine learning approaches is measured in order to assess their performance.