Statistics, Machine Learning & Citizen Science: Structures & Vocabulary

Modern astronomy produces datasets too large for manual inspection alone, making statistics, machine learning, and carefully designed citizen-science projects important parts of discovery. These methods assist classification and anomaly detection but do not replace measurement validation or physical interpretation.

Machine-learning models learn statistical relationships in training data and then predict labels or properties for new data. Their output can inherit selection biases, labeling errors, or distribution shifts, so performance must be evaluated on representative independent data.

Automated pipelines classify transients, galaxies, spectra, and exoplanet candidates at survey scale. Human-assisted projects have also identified unusual objects that automated filters initially missed.

A pipeline may include features, labels, training/validation/test sets, probabilistic classification, anomaly scores, uncertainty calibration, and human review. Citizen science adds structured human pattern recognition.