Search

Raman and infrared spectroscopy, used as individual and low-level fused datasets, were evaluated to identify and quantify the presence of adulterants (palm oil, PO; ω-3 concentrates in ethyl ester, O3C and fish oil, FO) in krill oil. These datasets were qualitatively analysed with principal component analysis (PCA) and classified as adulterated or unadulterated using support vector machines (SVM). Using partial least squares regression (PLSR), it was possible to identify and quantify the adulterant present in the KO mixture. Raman spectroscopy performed better (r2 = 0.98; RMSEP = 2.3%) than IR spectroscopy (r2 = 0.91; RMSEP = 4.2%) for quantification of O3C in KO. A data fusion approach further improved the analysis with model performance for quantification of PO (r2 = 0.98; RMSEP = 2.7%) and FO (r2 = 0.76; RMSEP = 9.1%). This study demonstrates the potential use of Raman and IR spectroscopy to quantify adulterants present in KO. [ABSTRACT FROM AUTHOR]

Herbal-based products are becoming a widespread production trend among manufacturers for the domestic and international markets. As the production increases to meet the market demand, it is very crucial for the manufacturer to ensure that their products have met specific criteria and fulfil the intended quality determined by the quality controller. One famous herbal-based product is herbal tea. This paper investigates bio-inspired flavour assessments in a data fusion framework involving an e-nose and e-tongue. The objectives are to attain good classification of different types and brands of herbal tea, classification of different flavour masking effects and finally classification of different concentrations of herbal tea. Two data fusion levels were employed in this research, low level data fusion and intermediate level data fusion. Four classification approaches; LDA, SVM, KNN and PNN were examined in search of the best classifier to achieve the research objectives. In order to evaluate the classifiers’ performance, an error estimator based on k-fold cross validation and leave-one-out were applied. Classification based on GC-MS TIC data was also included as a comparison to the classification performance using fusion approaches. Generally, KNN outperformed the other classification techniques for the three flavour assessments in the low level data fusion and intermediate level data fusion. However, the classification results based on GC-MS TIC data are varied.

Herbal-based products are becoming a widespread production trend among manufacturers for the domestic and international markets. As the production increases to meet the market demand, it is very crucial for the manufacturer to ensure that their products have met specific criteria and fulfil the intended quality determined by the quality controller. One famous herbal-based product is herbal tea. This paper investigates bio-inspired flavour assessments in a data fusion framework involving an e-nose and e-tongue. The objectives are to attain good classification of different types and brands of herbal tea, classification of different flavour masking effects and finally classification of different concentrations of herbal tea. Two data fusion levels were employed in this research, low level data fusion and intermediate level data fusion. Four classification approaches; LDA, SVM, KNN and PNN were examined in search of the best classifier to achieve the research objectives. In order to evaluate the classifiers' performance, an error estimator based on k-fold cross validation and leave-one-out were applied. Classification based on GC-MS TIC data was also included as a comparison to the classification performance using fusion approaches. Generally, KNN outperformed the other classification techniques for the three flavour assessments in the low level data fusion and intermediate level data fusion. However, the classification results based on GC-MS TIC data are varied. [ABSTRACT FROM AUTHOR]