A Survey on Implementing Machine Learning Algorithm to Predict Diabetes Stages and Preventing Elevated Blood Glucose Levels

Abstract

Diabetes, a chronic metabolic disorder, A health condition with rising prevalence. Accurate prediction of diabetes stages and proactive management of elevated blood glucose levels are crucial for effective treatment and prevention of complications. This study investigates the implementation of machine learning algorithms, specifically Support Vector Machine (SVM), Random Forest, and k-Nearest Neighbors (KNN), to address these critical aspects of diabetes care. In this research, a comprehensive data-set comprising clinical and demographic information of individuals was utilized. Data preprocessing techniques, including feature selection, normalization, and handling of missing values, were employed to prepare the data-set for modeling. These Algorithms were trained and evaluated for their effectiveness in predicting diabetes stages and identifying individuals at risk of elevated blood glucose levels. The previous techniques used in diabetes prediction performs with moderate accuracy and optimization which will not sufficient to attain maximum level of prediction.