ANALISIS FAKTOR RISIKO SINDROM OVARIUM POLIKISTIK (PCOS) PADA WANITA USIA SUBUR MENGGUNAKAN ALGORITMA EXPLAINABLE AI: POHON KEPUTUSAN VS RANDOM FOREST BERBASIS DATA KLINIS: RISK FACTOR ANALYSIS OF POLYCYSTIC OVARY SYNDROME (PCOS) IN WOMEN OF REPRODUCTIVE AGE USING EXPLAINABLE AI ALGORITHMS: DECISION TREE VS RANDOM FOREST BASED ON CLINICAL DATA

Muhammad Syaifur Rohman (Universitas Dian Nuswantoro)
Tsalisa Noor Rahmawati (Universitas Muhammadiyah Semarang)

RABIT: Jurnal Teknologi dan Sistem Informasi Univrab
Vol 10 No 2 (2025): Juli

Polycystic Ovary Syndrome (PCOS) is a common endocrine disorder affecting women of reproductive age and is a major cause of infertility and metabolic disturbances. This study aims to analyze risk factors of PCOS using Decision Tree and Random Forest classification algorithms based on clinical data. The dataset consists of 541 female patient records with 11 clinical features including age, body mass index (BMI), hormonal levels, and physical symptoms such as hair growth and pimples. The models were evaluated using accuracy, precision, recall, and f1-score. The Decision Tree model achieved an accuracy of 83.33%, with hair growth and weight gain as dominant features, while the Random Forest model achieved 82.41% accuracy and showed better performance in detecting positive cases (recall = 63.64%). The findings highlight that a combination of hyperandrogenism symptoms and hormonal indicators are key predictors of PCOS. This study demonstrates that machine learning algorithms can serve as effective tools to support early diagnosis of PCOS using clinical data