MAGIC Gamma Telescope

OBJECTIVE • To analyze the MAGIC Gamma Telescope dataset to classify gamma-ray sources from background cosmic radiation using machine learning techniques. • To improve detection accuracy, enabling better identification of high energy astrophysical phenomena. DATA DESCRIPTION The MAGIC Gamma Telescope dataset contains data on 19,020 instances with 11 attributes, including both predictive features and a binary classification target. • Number of instances: 19,020 • Attributes: 10 predictive features and 1 target (gamma or hadron classification) DATA FIELD • Predictive Features: Statistical and physical measurements (e.g., energy, length, width) from the telescope's observations. • Target Variable: Binary classification indicating whether an observation corresponds to gamma rays or hadrons (background cosmic rays). TARGET AUDIENCE • Astrophysicists studying high-energy cosmic phenomena. • Data scientists working on classification and predictive modeling in physics. • Engineers developing advanced telescopes and observational systems. END USER • Research institutions and universities conducting astrophysical research. • Space agencies and observatories managing gamma-ray detection equipment. FEATURES OF PROJECT • Preprocessing and cleaning of high-dimensional scientific data. • Development of a machine learning model to classify gamma rays. • Visualization of feature importance and data distribution. • Evaluation of model performance using metrics like precision, recall, and F1-score. PROJECT SCOPE • Identifying high-energy cosmic events with improved classification accuracy. • Exploring feature engineering techniques to enhance model performance. • Demonstrating the application of AI/ML in astrophysics research. PROJECT LIMITATION • The dataset is limited to observations from a single telescope, potentially affecting generalization. • Assumes balanced class distribution, though real-world data may have imbalances. • Requires advanced computational resources for training complex models. OUTCOME • A trained machine learning model capable of classifying gamma rays with high precision. • Insights into the physical features most critical for distinguishing gamma rays. • Contribution to astrophysical research by enhancing gamma-ray detection methods.

Built With

• jupitor
• ml
• python