Data fusion of electronic senses and near-infrared spectroscopy for accurate authentication and quality evaluation of Hippophae species.

Sea buckthorn (Hippophae spp.) is a valuable medicinal and edible plant whose quality is highly dependent on its origin. Rapid species classification is therefore essential for quality assurance of sea buckthorn. This study developed an integrated method combining electronic eye (E-eye), electronic tongue (E-tongue), and near-infrared spectroscopy (NIRS) with machine learning for the rapid classification and quality evaluation of three Hippophae species. Individual models based on E-eye, E-tongue, and NIRS showed discriminative ability. Data fusion achieved 100.00% independent validation accuracy for all classifiers, with cross-validation accuracies of 100.00% for SVM and KNN and 94.74% for DT. Quantification of flavonoid glycosides revealed that H. rhamnoides subsp. sinensis contained the highest levels, while assessment of antioxidant activities showed that H. gyantsensis exhibited the strongest activity. The proposed data fusion strategy provides a rapid, accurate, and cost-effective solution for authentication and quality control of Hippophae, with promising applications in food and pharmaceutical industries.