Machine learning classifiers distinguished between Cannabis sativa and Cannabis indica cultivars based on user-reported flavours and subjective effects with high accuracy. Analysis of a large dataset from Leafly.com and chemical composition data from Psilabs.org revealed significant correlations between terpene and cannabinoid content and subjective effect and flavour tags. Reported effects clustered into three groups: unpleasant, stimulant, and soothing. Terpene profiles matched user perceptual characterizations, particularly for terpene-flavours associations. The findings suggest that flavour perception could serve as a reliable marker to indirectly characterize cannabis psychoactive effects, as terpene content is robustly inherited and less influenced by environmental factors.
Machine learning analysis of a large public dataset where users freely reported their experiences with cannabis strains, combined with chemical composition data, reveals that cannabis strains can be reliably classified into three major clusters corresponding to Cannabis sativa, Cannabis indica, and hybrids based on self-reported effect and flavor tags. Terpene profiles matched users' perceptual characterizations and could predict associations between different psychoactive effects, while cannabinoid content was variable even within individual strains. The findings suggest that flavor perception, reflecting robustly inherited terpene content, could serve as a reliable marker to predict psychoactive effects, offering a data-driven approach to strain classification for the growing medicinal and recreational cannabis market.