Prediction of Diabetes Using Machine Learning: Analysis of 70,000 Clinical Database Patient Record

Abstract

Diabetes is a disease that occurs when the blood glucose level or blood sugar level meets high values. The level of sugar is increased in the human body is due to several factors like obesity, physical inactivity, gender, age, family history, food habits, and so on. Based on these attributes and with the help of machine learning techniques one can foresee diabetes. According to the increasing morbidity, the number of patients who suffer from diabetes will reach 642 million in 2040, which indicates one of the ten adults in the world will suffer from diabetes. Algorithms that are used in Machine learning can apply in the various medical health field to detect and predict diseases. In this paper, we applied Logistic Regression, Random Forest, Support Vector Machine (SVM), and K-Nearest Neighbor (KNN) machine learning algorithms to predict diabetes in Diabetes 130-US hospitals for the years 1999-2008 Data Set and Pima Indian Diabetes Dataset. We made a comparative study of the accuracy of all machine learning algorithms. In our diabetic prediction model, we got a higher accuracy value for the random forest algorithm.

Ähnliche Arbeiten