Analisis Kinerja Ensemble Learning dan Algoritma Tunggal dalam Klasifikasi Sindrom Ovarium Polikistik Menggunakan Random Forest, Logistic Regression, dan XGBoost

Djaka, Thesa Permatasari Djaka
Nurul Anisa Sri Winarsih

Infotekmesin
Vol 16 No 1 (2025): Infotekmesin: Januari 2025

Polycystic ovary syndrome (PCOS) is a hormonal disorder that is the most common cause of anovulation and infertility in women of reproductive age, affecting approximately 5-10% of the population, with up to 70% of cases undiagnosed. This highlights the need for early detection methods with high accuracy for timely treatment. Previous research utilized a classification method based on the K-Nearest Neighbor (KNN) algorithm, which demonstrated good performance with an accuracy of 93%, precision of 100%, recall of 82%, and F1-Score of 90%. This study proposes using an ensemble learning method with a voting classifier technique that combines several classification models: Random Forest Classifier, Logistic Regression, and XGBoost Classifier. The results show that the proposed method performs better with an accuracy of 95%, precision of 100%, recall of 85%, F1-Score of 92%, and an AUC (Area Under Curve) value of 94.34%