Comparison of Different Machine Learning Models for diabetes detection

Abstract

Diabetes metilus which is commonly known as diabetes is a major metabolic disorder which has a severe effect on a human being. Diabetes results in high blood sugar. In a human body, there is a hormone which is secreted by the pancreas called insulin which helps to move the glucose from the blood to the cells which are used for energy later. In diabetes, one's body doesn't produce insulin inadequate amount or is impotent to use insulin effectively. When diabetes is not treated properly the danger of heart attack, retinopathy or vision loss, skin conditions and some other disorders increases. There are more than a million people currently who are suffering from this disease. The detection of diabetes in early stages can help one to take appropriate measures. The rapid increase in the number of people suffering from diabetes is gaining everyone's attention. The subset of artificial intelligence is Machine learning(ML) in which the system learns from the experience without doing any explicit programming. In this research, we have applied the machine learning technique for the detection of patterns and risk factors in Pima Indian diabetes dataset using python data manipulation tool. For the categorization of the patient into diabetic or non-diabetic, we have applied six machine learning algorithms specifically support vector machine(SVM), k-nearest neighbour (KNN), Gradient boosting, Decision tree, Random forest and logistic regression.

Ähnliche Arbeiten