Optimization Accuracy of Diabetes Disease Prediction using Gradient Boosting Machine Learning Technique

Abstract

Diabetes is one of the most prevalent chronic diseases worldwide, affecting millions of individuals and leading to severe health complications if not diagnosed and managed at an early stage. Accurate prediction of diabetes can assist healthcare professionals in making timely decisions and improving patient outcomes. This study presents an optimized diabetes disease prediction system using the Gradient Boosting Machine Learning Technique. The proposed approach utilizes patient health parameters such as glucose level, blood pressure, body mass index (BMI), insulin concentration, age, and other clinical attributes to predict the likelihood of diabetes. Data preprocessing techniques, including missing value handling, normalization, and feature selection, are applied to enhance data quality and model performance. The Gradient Boosting algorithm is employed due to its ability to combine multiple weak learners into a strong predictive model, thereby improving classification accuracy and reducing prediction errors. Experimental results demonstrate that the proposed model achieves superior performance compared to conventional machine learning classifiers in terms of accuracy, precision, recall, and F1-score. The optimized Gradient Boosting-based prediction system provides an effective and reliable tool for early diabetes detection, supporting healthcare practitioners in preventive diagnosis and personalized treatment planning.