VASANTHAKUMAR P

Tools & Technologies: Python, Pandas, Scikit-learn, Streamlit, HTML, Excel

Description:
Developed a machine learning-based web application to predict the number of calories burned during physical activities based on user-specific inputs such as age, gender, height, weight, heart rate, body temperature, and type of activity. The project aimed to offer real-time, personalized insights into calorie expenditure to promote health and fitness awareness.

Key Contributions:

• Collected and cleaned a dataset of 10,000+ records covering various exercise types.
• Applied feature engineering techniques to improve model accuracy.
• Trained and evaluated multiple models (Random Forest, XGBoost), achieving high prediction accuracy.
• Integrated the trained model into an interactive Streamlit web app for user-friendly predictions.
• Enhanced the dataset with MET-based formula calculations for more realistic calorie estimations.
• Focused on scalability and usability, enabling support for multiple workout types and durations (up to 5 hours).

Outcome:
Successfully built a predictive system capable of assisting users in understanding energy expenditure, useful for fitness tracking applications and personalized health monitoring.