A Local Dwarf Galaxy Search Using Machine Learning

Key Points

• A machine learning classification model achieves 95% precision in identifying local, low-mass galaxies, enhancing galaxy catalogs.
• The model confirms low-mass galaxy candidates with various spectra, yielding independent precisions of approximately 100%, 96%, and 97%.
• By analyzing combined photometric data from numerous surveys, researchers can effectively separate low-mass galaxy signals from contaminants.
• Features encoding surface brightness are crucial for accurate classifications, indicating their significance in galaxy detection processes.

Abstract