KLASIFIKASI RISIKO OBESITAS BERBASIS GRADIENT BOOSTING PADA DATA MEDIS

Abstract

Obesity is a growing health concern that can lead to various chronic diseases, making accurate risk identification an important preventive effort. The development of machine learning techniques enables the utilization of medical data to support intelligent decision-making in the healthcare domain. This study aims to apply the Gradient Boosting algorithm as a classification method to predict obesity risk based on medical data. The dataset used contains information related to eating habits, physical activities, and individual characteristics. The research process includes data preprocessing, data transformation and normalization, class mapping, and data partitioning into training and testing sets with a ratio of 70:30. The Gradient Boosting model is constructed using multiple decision trees with specific parameter settings to classify obesity risk into two categories, namely obese and non-obese. Model performance is evaluated using accuracy, precision, recall, and F1-score metrics. The experimental results show that the proposed model achieves good classification performance with an accuracy exceeding 90%, while the performance gap between training and testing data remains relatively small. This indicates that the model has strong generalization capability and does not suffer from overfitting. Therefore, the application of Gradient Boosting on medical data proves to be an effective approach for obesity risk classification and has the potential to support intelligent health information systems in assisting medical practitioners with more precise obesity prevention and management strategies.