MagicTelescope: Monte Carlo Simulated Gamma-Ray Telescope Events for Classification

Monte Carlo generated data simulating the registration of high-energy gamma particles in a ground-based atmospheric Cherenkov telescope. The dataset was created for a classification benchmark on numerical features, using the Corsika simulation program. It was donated by P. Savicky from the Institute of Computer Science, AS of CR, Czech Republic, and originates from the UCI repository.

Use Cases

• Binary classification of gamma-ray signals from cosmic-ray background based on Hillas ellipse parameters.
• Feature engineering and analysis for pattern recognition in simulated particle shower images.
• Benchmarking machine learning algorithms for classification tasks on purely numerical features.
• Studying the statistical discrimination of elongated cluster patterns in telescope camera data.

Strengths

• Data is generated via a documented Monte Carlo simulation program (Corsika), ensuring controlled and reproducible conditions.
• Dataset is part of a recognized tabular data benchmark, suggesting a standardized transformation and evaluation context.
• Includes a clear citation to the original scientific publication describing the event classification methods.

Limitations

• Row count, column definitions, and sample data are unknown, which limits suitability assessment and requires inspection after download.
• Last update date is unknown; freshness of the data is unverified.
• The description focuses on the scientific context; specific data schema and feature semantics must be inferred from the downloaded files.

Provenance

Source

UCI Machine Learning Repository, originally from the MAGIC telescope project.

Collection Method

Monte Carlo simulation using the Corsika program.

Time Range

null

Freshness

null

Geography

null