Answers are grounded in this paper and cite their sources.
The authors have developed an artificial neural network model using odor threshold (OT) property data for diverse odorant components present in black tea (76 components) and coffee (46 components). The models were validated in terms of both internal and external validation criteria signifying acceptable results. The authors found the significant features controlling the OT property using Mean Absolute Error (MAE)-based criteria in a backward elimination of descriptors, one in each turn. The present results well-corroborated the previously published PLS-regression based chemometric model results.
No takes yet. Share an insight, caveat, or question.
Krishna et al. (2019) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: