[Subset] Unique Identification of 50,000+ Virtual Reality Users from Head & Hand Motion Data

A sub-sampled version of the main experiment from "Unique Identification of 50,000+ Virtual Reality Users from Head & Hand Motion Data," with up to 5,000 users instead of 50,000. Allows for a partial reproduction of the results with significantly fewer resources than running the main evaluation. To run, run C1.py, C2.py, and C3.py. After running C1.py, an extrapolation of the results will be printed in the terminal and saved as results.pdf. After running C2.py, the feature importance distribution will be printed in the terminal and saved as features.pdf. After running C3.py, the open-world evaluation results will be printed in the terminal.

The following dependencies are required:

• PyTorch (torch) v1.13.1
• pandas v1.5.2
• tqdm v4.64.1
• scikit-learn (sklearn) v1.2.0
• NumPy (numpy) v1.24.0
• LightGBM (lightgbm) v3.3.3
• Joblib (joblib) v1.2.0
• NetworkX (networkx) v3.0
• Matplotlib (matplotlib) v3.6.2