MAGIC Gamma Telescope: Simulated High-Energy Particle Events for Classification

Monte Carlo-generated data from 2004 simulating the registration of high-energy gamma particles in a ground-based atmospheric Cherenkov telescope. The dataset contains 19,020 instances with 11 attributes describing shower image parameters, generated using the Corsika simulation code. It is intended for statistical discrimination between gamma-ray signals and hadronic cosmic-ray background.

Use Cases

• Binary classification of gamma-ray signals versus cosmic-ray background based on Hillas parameters.
• Feature importance analysis for shower image characteristics like ellipse major/minor axis and asymmetry.
• Training models for particle identification in high-energy physics experiments using simulated data.
• Benchmarking classification algorithms on a physics-based, structured tabular dataset.

Strengths

• Contains 19,020 simulated instances, providing a substantial sample for model training.
• Includes 11 attributes with clear physical interpretations, such as ellipse axes and pixel concentration ratios.
• Data generation is documented with a reference to the established Corsika Monte Carlo simulation code.

Limitations

• Data is simulated from 2004; its correspondence to real-world observations is unverified.
• Column-level documentation is incomplete; the description cuts off for one attribute (fM3Long).
• The last update date and original data freshness are unknown.

Provenance

Source

Generated via the Corsika Monte Carlo simulation program.

Collection Method

Monte Carlo simulation to model extensive air showers initiated by high-energy gamma rays.

Time Range

Simulation data from 2004.