Perbandingan Kinerja Algoritma Random Forest, AdaBoost, dan Gradient Boosting dalam Memprediksi Risiko Penyakit Hipertensi

Abstract

Hypertension disease risk prediction is one of the challenges in the health field that can be supported by the development of machine learning models. Hypertension is a chronic condition that can lead to various serious complications, such as heart disease and stroke, so early detection is very important. However, conventional methods of diagnosing hypertension often require extensive medical examinations and are not always accessible to all individuals. Therefore, the development of artificial intelligence-based predictive models can be a more efficient solution in supporting the early detection of hypertension.This study aims to compare the performance of three popular machine learning algorithms, namely Random Forest, AdaBoost, and Gradient Boosting, in predicting hypertension risk. The most effective algorithm will be used in future research for program development. The dataset used consists of relevant medical and demographic data, such as blood pressure, body mass index, age, gender, and family history of hypertension. The model is built using a supervised learning approach, where the data is labeled based on the patient's hypertension condition. Evaluation was conducted using accuracy, precision, recall, and F1-score metrics to assess the performance of each algorithm.The methods used in this research include data preprocessing, feature selection, model training, and model performance evaluation. In addition, this research also designs an artificial intelligence-based hypertension prediction application that is expected to provide recommendations to users based on the model's prediction results.The results of this research are expected to provide insight into the most effective machine learning algorithms in hypertension risk prediction, considering the trade-off between accuracy and computational efficiency. Hypothesized based on previous research, Random Forest algorithm is better than the other two algorithms.