Researchers at the National University of Singapore, in collaboration with the Shanghai Institute of Technology, have deployed machine learning to systematically identify plant-derived scent compounds with sleep-promoting properties. Led by Assistant Professor Zhang Dachuan, the study compiled a curated dataset of 2,391 aroma molecules sourced from 991 aromatic plant species. This comprehensive library was used to train an AI model designed to recognize chemical patterns associated with sleep-enhancing activity. The model demonstrated a testing accuracy of 96.1 percent, effectively distinguishing bioactive sleep compounds from inactive ones. Based on these predictions, the team isolated five commercially available molecules for experimental validation. Four of these compounds, carvacrol, safranal, vanillin, and methyl eugenol, significantly reduced wakefulness and enhanced non-rapid eye movement sleep. Further analysis revealed that these molecules interact with gamma-aminobutyric acid receptors, the primary calming signaling pathway targeted by conventional prescription sleep medications. Beyond individual molecules, the research mapped the chemical density of sleep-promoting compounds across plant families. Asteraceae, Lamiaceae, and Lauraceae emerged as the most prolific sources, with species such as lavender and perilla flagged as high-priority candidates for future development. The findings, recently featured on the cover of Digital Discovery, establish a data-driven framework for translating traditional botanical knowledge into evidence-based wellness applications. Rather than presenting an immediate commercial product, the study delivers a practical chemical roadmap that accelerates the discovery of natural ingredients for functional foods, therapeutic fragrances, and targeted sleep supplements. Looking ahead, the research team will investigate the long-term safety profiles of these compounds, evaluate how molecular mixtures interact within biological systems, and expand validation into broader human clinical trials. By bridging artificial intelligence with ethnobotanical chemistry, this initiative demonstrates a scalable approach to developing safer, plant-based alternatives to traditional pharmaceutical sleep aids.