An Intelligent Web Platform for Menstrual Tracking, Hygiene, and Community Support Using Machine Learning

Abstract

Menstrual health is crucial for women's well-being, yet many digital health solutions fall short. Most period-tracking applications use static calendar-based algorithms that do not adapt to irregular cycles, leading to inaccurate predictions and limited support. They often fail to provide comprehensive features like personalized reminders and expert guidance, making them less effective during unexpected challenges. This paper proposes an intelligent menstrual health web application that utilizes machine learning for accurate period predictions, offers real-time reminders for changing menstrual products, and helps users locate nearby washrooms and gynecologists. The platform will include a chatbot for user queries, a blog for sharing stories, a community chat, and a “Craving Decoder” feature for healthier snack alternatives. Built with Python (Flask) for the backend and HTML, CSS, and JavaScript for the frontend, this system aims to improve prediction accuracy and user support through a complete suite of features. Performance evaluations indicate that Decision Tree and Random Forest models deliver the most balanced results in accuracy and minority class detection.