Purpose <p>Previous studies have primarily relied on traditional methods and lacked a comprehensive framework to evaluate the psychological and lifestyle factors associated with sleep quality in PCOS patients. The study’s goal is to use machine learning to identify the predictive factors that show a relationship with sleep quality in women with PCOS.</p> Methods <p>The study used a cross-sectional design with female students and staff as participants from January to June 2023. The Hospital Anxiety Depression Scale (HADS) and Pittsburgh Sleep Quality Index (PSQI) were used to quantitatively assess psychological status and sleep quality. Data was analyzed using a decision tree model to find how lifestyle choices and psychological health are related to sleep patterns. The dataset was split into 70% training and 30% testing sets. To address class imbalance (64 PCOS vs. 141 non-PCOS), random oversampling and basic sampling was applied to the training data.</p> Results and Discussions <p>This study included 205 respondents, with 64 diagnosed with PCOS and 141 without PCOS. Women with PCOS had poorer sleep quality overall (<i>P</i> &lt; 0.05) than those in the control group. Physical activity was the lifestyle variable that had the most significant negative predictor of PCOS patients’ sleep quality, while anxiety was the psychological variable with the strongest association with sleep quality using the decision tree model. The accuracy of the model is 93.9% as well.</p> Conclusion <p>Our research revealed that the quality of sleep experienced by women with PCOS is significantly associated with anxiety and physical activity. These findings indicate that targeted therapies concentrating on anxiety management and lifestyle change may be beneficial in improving sleep quality in women with PCOS. Additionally, the application of interpretable machine learning models, such as decision trees, could help in the development of clinical screening tools to enhance sleep and overall care in these individuals.</p>