A machine-learning-driven dataset of 144,111 polycyclic aromatic hydrocarbon infrared spectra for interpreting aromatic infrared bands
Researchers created a massive machine-learning dataset of 144,111 infrared spectra for polycyclic aromatic hydrocarbons (PAHs). This new database uses a neural network framework to predict spectral features from molecular structures. The goal is to provide a much more complete reference library than existing ones, helping scientists interpret astronomical observations of interstellar molecules.