Effective Prediction of Diabetes Mellitus using Nine different Machine Learning Techniques and their Performances

Shashank Joshi,Neha Sagar,Dept Of Computer Engg, Vit, Pune, India ,Anamika Rathod,Vijayendra Gaikwad,Sairam Rathod

doi:10.35940/ijeat.e9626.069520

Shashank Joshi, Neha Sagar + Show 4 more

Open Access

https://doi.org/10.35940/ijeat.e9626.069520

Copy DOI

Abstract

Diabetes is a disease where the predominant finding is high blood sugar. The high blood sugar may either be because of deficient insulin production (Type 1) or insulin resistance in peripheral tissue cells (Type 2). Many problems occur if diabetes remains untreated and unidentified. It is additional inventor of various varieties of disorders for example: coronary failure, blindness, urinary organ diseases etc. Nine different machine learning techniques are used in this research work for prediction of diabetes. A dataset of diabetic patient’s is taken and nine different machine learning techniques are applied on the dataset. Positive likelihood ratio, Negative likelihood ratio, Positive predictive value, Negative predictive value, Disease prevalence, Specificity, Precision, Recall, F1-Score ,True positive rate, False positive rate of the applied algorithms is discussed and compared. Diabetes is growing at an increasing in the world and it requires continuous monitoring. To check this we use Logical regression, Random forest, Logical regression CV, Support Vector Machine, Artificial Neural Network (ANN), Decision Tree, k-nearest neighbors (KNN), XGB classifier

Highlights

The annual report of World Health Organization, add up to the number of individuals experiencing diabetes is estimated to be 9.3% (463 million people), rising to 10.2% (578 million) by 2030 and 10.9% (700 million) by 2045
Diabetes affects the ability of the body in producing the hormone insulin or increasing the resistance of body cells to the insulin produced, which in turn makes the metabolism of carbohydrate abnormal and raise the levels of glucose in the blood
Umatejaswi has presented the algorithms like Decision Tree, Support Vector Machine (SVM), Naive Bayes for identifying diabetes using data mining techniques [6]

Summary

INTRODUCTION

The annual report of World Health Organization, add up to the number of individuals experiencing diabetes is estimated to be 9.3% (463 million people), rising to 10.2% (578 million) by 2030 and 10.9% (700 million) by 2045. Type diabetes occurs because of the failure of pancreas to supply enough hypoglycemic agent i.e. insulin. This type is labelled as "juvenile diabetes". The type one polygenic disease found in children beneath twenty years old People suffer throughout their life because of the type one diabetes. Type 2 diabetes is an adult-onset diabetes value occurring in obese individual or those with the family history. In this there is a resistance of muscles, fat and other tissues to insulin. Type known as Gestational diabetes occurs when a woman is pregnant and develops the high blood sugar levels without a previous history of diabetes. From the above twelve measures which machine learning technique is the best for prediction of diabetes is calculated

LITERATURE REVIEW

PROPOSED MODEL

RESULTS

CONCLUSION