Nutrition for women largely changes based on their current menstrual phase due to differences in energy levels, hormones, and other health factors. To address this, the proposed work aims to help the menstruator accurately predict their menstrual phases and have comprehensive access to the required and recommended nutrition for it. This work presents a machine learning approach for predicting menstrual cycle phases and providing personalized dietary recommendations, focusing on utilizing the XGBoost algorithm with SMOTE and hyperparameter tuning. The model processes user-input data based on their menstrual cycle to accurately predict phases of the regular menstrual cycles with an average accuracy of 99.39%. XGBoost was selected for its superior performance in handling imbalanced datasets, which is crucial for effectively predicting less frequent phases such as Ovulation. By leveraging a phase-specific approach, the system ensures that nutritional advice is aligned with the user’s unique cycle patterns, improving overall well-being. The unique approach of combining accurate phase prediction with personalized nutrition recommendations ensures actionable insights for users. This work demonstrates the potential for real-world applications in personalized healthcare, particularly in managing menstrual health and nutrition more effectively.