Enhancing polycystic ovary syndrome classification accuracy using XGBRF model
Sonam Junejal, Supreet Sainil, Reema Goya, Souvik Maiti, Vinay Thakur · 2025
Polycystic ovary syndrome (PCOS) is a common endocrine condition that affects women in their reproductive years all over the world. It is typified by irregular menstruation, hormone abnormalities, and ovarian cysts. Prompt and precise identification of PCOS is essential for efficient therapy and avoidance of related health is-sues. In this study, we employ cutting-edge machine learning techniques to present a novel model for the precise diagnosis of PCOS. Our model incorporates a range of clinical and biochemical factors that are frequently linked to polycystic ovary syndrome (PCOS), such as menstrual patterns, anthropometric; and hormone levels. We optimized the hyperparameters of two well-known classifiers, Random Forest and XGBoost, using exhaustive grid search approaches. While learning rate, maximum tree depth, and regularization terms were set for XGBoost, parameters like the number of trees, maximum depth, and minimum samples for node splitting were optimized for Random Forest.