Classification of Diabetes Using Machine Learning Technics

Abstract

Diabetes is one of the most widespread chronic diseases worldwide. More than 400 million people in the world suffer from it, and their number is increasing significantly every year. The continents of the world, especially Asia and Africa, are experiencing a high rate of infection due to a number of factors and environmental factors. Related to diet and healthy habits used in many developing countries. Diabetes is considered one of the most serious chronic diseases that affect the human body due to impaired secretion of the hormone insulin, because the pancreas is unable to secrete a sufficient amount of insulin, or when the cells of the human body do not accept insulin and don't use it. Effective, causing a sudden drop or rise in blood sugar levels. It has a huge impact on human health and type 2 diabetes is the most common form in humans as many factors contribute to its spread, such as genetic factors, poor diet and lack of regular exercise. In this paper we used the machine learning technical to classification of diabetes such as logistic regression, random forest, decision tree, vector classification, KNN and Naïve bayes and the accuracy was 77.66%, 76.33%, 98.66%, 75.33%, 80.66% and 80% The rest of the general standards used in classification are mentioned in the results of this paper.