AI‑Driven Statistical Analysis of Dietary Habits and Lifestyle Factors Associated with PCOS Among Women

Polycystic Ovary Syndrome (PCOS) is one of the most common endocrine disorders affecting women of reproductive age and is closely associated with metabolic disturbances, hormonal imbalance, and lifestyle factors such as diet and stress. This study investigates the relationship between dietary habits, nutritional awareness, lifestyle behaviors, and PCOS symptoms using advanced statistical and machine learning techniques. The dataset consists of 500 observations including variables such as age group, dietary patterns, physical activity, stress levels, menstrual health indicators, and nutritional awareness. Statistical analyses including correlation analysis and survival modeling were applied alongside machine learning techniques such as Principal Component Analysis (PCA), K-Means clustering, Random Forest, Logistic Regression, Bayesian models, and deep learning approaches. The MLP model showed effective classification performance, and SHAP analysis enhanced interpretability The results demonstrate that stress frequency, dietary habits, and menstrual irregularities significantly influence PCOS risk. Ensemble machine learning models showed higher predictive accuracy compared to traditional statistical models. Clustering analysis revealed distinct patient groups based on lifestyle behaviors. The findings highlight the importance of lifestyle interventions, improved nutritional awareness, and AI-driven predictive modeling for early detection and management of PCOS among women.