AI-Based Smart Fitness Tracker with Real Time Activity Prediction Personalized Health Monitoring

Prof. Rohan B. Kokate

Department of Computer Application J. D. College of Engineering & Management, Nagpur, India

Prof. Aman Singh

Department of Computer Application J. D. College of Engineering & Management, Nagpur, India

Shreyash Anil Dange

Department of Computer Application J. D. College of Engineering & Management, Nagpur, India

Abstract:Modern lifestyles demand intelligent systems that can monitor physical activity and provide personalised fitness insights. Traditional fitness trackers often rely on basic metrics like step counting, which limits their ability to accurately classify activities or offer meaningful recommendations. This paper presents an AI-based smart fitness tracker that predicts activities such as walking, running, cycling, and gym workouts using inputs like heart rate, speed, activity level, and user weight. The system employs lightweight machine learning-inspired logic for real-time activity prediction, calorie estimation, and performance monitoring. It also includes user profile management and workout history tracking to enhance personalisation. Experimental results demonstrate accurate activity classification with efficient performance, making the system a scalable, user-friendly, and effective solution for modern fitness monitoring applications.Index Terms— Fitness Tracking, Activity Prediction, Artificial Intelligence, Machine Learning, Health Monitoring, Calorie Estimation, Real-Time Prediction, Web-Based Application