COMPARATIVE STUDY OF MACHINE LEARNING ALGORITHMS CHINE LEARNING ALGORITHMS FOR DIABETES RISK PREDICTION: A FOCUS ON K-NN, SVM, DECISION TREE, AND RANDOM FOREST.

Abstract

The global health situation is significant due to diabetes mellitus, a chronic metabolic disorder. One of the chronic diseases known as chronic metabolic ailment is caused by persistently elevated blood sugar levels. It is believed to be among the deadliest illnesses worldwide. If an accurate early prognosis is available, the severity and risk factors of diabetes can be greatly reduced. The early diagnosis of diabetes can be aided by algorithms for machine learning. Early identification can help diabetes patients reduce their health risks. The results can be beneficial for doctors, patients, and family members of patients. It is essential to estimate the patient’s state upon entry in order to allocate resources correctly in healthcare settings with limited resources. Medical diagnosis accuracy is increased and costs are decreased using machine learning approaches. This research paper presents a comprehensive comparative study of machine learning algorithms, namely k-Nearest Neighbours (k-NN), Support Vector Machine (SVM), Decision Tree, and Random Forest, to identify the most effective model for diabetes risk prediction. All four algorithms’ results are assessed using a range of metrics, including recall, Fmeasure, accuracy, and precision. The number of correctly and wrongly classified cases is used to calculate accuracy. The results demonstrate that when compared to other algorithms, Random Forest performs with the greatest accuracy of 99.03%. preceded by Decision Tree with an accuracy of 95% preceded by SVM with an accuracy of 90% and K-Nearest Neighbour with an accuracy of 89%.