2025ApJS..277...21Q

Interstellar molecules, which play an important role in astrochemistry, are identified using observed spectral lines. Despite the advent of spectral analysis tools in the past decade, the identification of spectral lines remains a tedious task that requires extensive manual intervention, preventing us from fully exploiting the vast amounts of data generated by large facilities such as the Atacama Large Millimeter/submillimeter Array. This study aims to address the aforementioned issue by developing a framework for automated line identification. We introduce a robust spectral fitting technique applicable to spectral line identification with minimal human supervision. Our method is assessed using published data from five line surveys of hot cores, including W51, Orion-KL, Sgr B2(M), and Sgr B2(N). By comparing the identified lines, our algorithm achieves an overall recall of ∼74%-93%, and an average precision of ∼78%-92%. Our code, named SPECTUNER, is publicly available on GitHub.