Letter to the Editor

Full-spectrum infrared fingerprinting: A transformative AI paradigm for interstellar polycyclic aromatic hydrocarbons

Laboratory for Relativistic Astrophysics, Department of Physics, Guangxi University, 530004 Nanning, China

^★ Corresponding author: This email address is being protected from spambots. You need JavaScript enabled to view it.

Received: 23 March 2026
Accepted: 25 May 2026

Abstract

Context. In the era of high-sensitivity infrared (IR) astronomy, traditional manual diagnostics are no longer sufficient to harvest the complex physical insights hidden within interstellar spectra.

Aims. We introduce a machine learning paradigm that bypasses the limitations of empirical band ratios by treating the complete IR spectrum of polycyclic aromatic hydrocarbons (PAHs) as a high-dimensional fingerprint.

Methods. Using a random forest classifier trained on ∼23 000 spectra, we achieved a robust F₁ score of ∼0.963 across 12 size and charge categories, maintaining high performance on unseen molecular mixtures.

Results. Interrogating the model’s decision-making process reveals that PAH size diagnostics are charge-dependent. Neutral PAHs are traced by C-H modes, while ionized species rely on 6 − 8 μm C-C morphology; however, the 12.5 μm feature remains a versatile tracer across multiple charge states.

Conclusions. This AI-driven paradigm offers a new route to interpret IR signatures and probe the chemical complexity of the interstellar medium.