Title: Prediction and diagnosis of diabetes using machine learning classifiers

Authors: Harsh Tyagi; Aditya Agarwal; Aakash Gupta; Kanak Goel; Anand Kumar Srivastava; Akhilesh Kumar Srivastava

Addresses: Department of Computer Science and Engineering, ABES Engineering College, Ghaziabad, UP, India ' Department of Computer Science and Engineering, ABES Engineering College, Ghaziabad, UP, India ' Department of Computer Science and Engineering, ABES Engineering College, Ghaziabad, UP, India ' Department of Computer Science and Engineering, ABES Engineering College, Ghaziabad, UP, India ' Department of Computer Science and Engineering, ABES Engineering College, Ghaziabad, UP, India ' Department of Computer Science and Engineering, ABES Engineering College, Ghaziabad, UP, India

Abstract: Major portion of diabetes in the world is of type 2 due to environmental conditions and lifestyle. If the diabetes is predicted at an early stage, it would really help in reducing its effects by the use of early medication. This article is based on machine learning model to predict diabetes based on diagnostic measurements. Machine learning can play an essential role in predicting presence/absence of diabetes mellitus (type 2 diabetes). The article presents the ML-based approach for prediction of the diabetes that makes use of algorithms like XGBoost, decision tree, random forest. In this, medical data of user is used as input and prediction of diabetes is done using mentioned algorithms. The output (prediction) will be the ensemble of the output of all three algorithms. That way, all the algorithms are used to make predictions and to establish a comparison between the accuracy obtained from these methods.

Keywords: machine learning; diabetes; random forest; XGBoot; decision tree.

DOI: 10.1504/IJFSE.2021.10044205

International Journal of Forensic Software Engineering, 2022 Vol.1 No.4, pp.335 - 347

Received: 06 Aug 2021
Accepted: 19 Oct 2021

Published online: 05 Jul 2022 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article