A machine learning approach fusing multisource spectral data for prediction of floral origins and taste components of Apis cerana honey.
Journal:
Food research international (Ottawa, Ont.)
PMID:
40263817
Abstract
This study explores the use of near-infrared (NIR), mid-infrared (MIR), and Raman spectral fusion for the rapid prediction of floral origins and main taste components in Apis cerana (A. cerana) honey. Feature-level fusion with the partial least squares regression - random forest (PLSR-RF) model achieved 100 % classification accuracy in identifying floral origins. Additionally, the model demonstrated strong predictive capability for sugars, amino acids, and organic acids, with R values ranging from 0.88 to 0.96, and performed exceptionally in predicting total organic acids and amino acids (R of 0.94 and 0.93, respectively). The PLSR-RF model showed effective clustering for proline, glucose, and fructose, achieving a 23.5 % improvement in predictive accuracy compared to data-level fusion. These findings confirm the efficacy of the PLSR-RF model for quantitative analysis of A. cerana honey.