Synthetic Triathlete Dataset for Injury Prediction Research (2024)

This dataset contains synthetic data for 1,000 triathletes over the year 2024 and is intended for use in injury prediction and athlete monitoring research. The data was generated to reflect realistic patterns observed in endurance athletes, while maintaining full privacy through synthetic generation techniques. The dataset includes three CSV files: athletes.csv – Contains static demographic and profile information for each synthetic athlete (e.g., age, gender, training background). daily_data.csv – Daily physiological and biometric readings collected via simulated wearable devices (e.g., heart rate, HRV, sleep). activity_data.csv – Timestamped records of individual training sessions (e.g., type, intensity, duration), each linked to an athlete ID. The dataset spans January 1, 2024 to December 31, 2024 and includes 366,000 daily records and 384,153 activity sessions. It is suitable for research on injury prediction, workload modeling, recovery analysis, and personalized training optimization. Citation: Please cite this dataset as:Rossi Leonardo. (2025). Synthetic Triathlete Dataset for Injury Prediction Research (2024). Zenodo. https://doi.org/10.5281/zenodo.15401061