This project was developed as part of a MATLAB hackathon challenge, with the goal of building a fitness tracker that works with real sensor data collected from a smartphone. The core idea was to use machine learning to recognize different physical activities (like sitting, walking, and running), then analyze them in terms of calories burned and steps taken using real-world GPS and accelerometer inputs.
The project was inspired by the challenge of making fitness tracking intelligent and interpretable, using only sensors available on a smartphone. Rather than relying on wearables or expensive hardware, the idea was to empower users to analyze their movement patterns through mobile data and simple visualizations. I was also motivated by the chance to combine concepts from machine learning, signal processing, and biomedical engineering in a single project.
Log in or sign up for Devpost to join the conversation.