Search citation statements
Paper Sections
Citation Types
Year Published
Publication Types
Relationship
Authors
Journals
Roasted ground coffees are targets of concern regarding intentional adulteration with cheaper foreign materials because, in this form, it may be difficult to detect due to the small particle size and the dark color. Therefore, a significant interest is developing fast, sensitive, and accurate methodologies to quantify adulterants in roasted coffees. This study investigated the potential of using near-infrared spectroscopy (NIR) to quantity robusta coffee and chicory in roasted arabica coffee. The adulterated arabica coffee samples were composed of robusta coffee or chicory ranging from 2.5 to 30% in increments of 2.5%. Four regression approaches were applied: gradient boosting regression (GBR), multiple linear regression (MLR), k-nearest neighbor regression (KNNR), and partial least squares regression (PLSR). The first three regression models were performed on the features extracted from linear discriminant analysis (LDA) or principal component analysis (PCA). Additionally, two classification methods were applied (LDA and KNN). The regression models derived based on LDA-extracted features presented better performances than PCA ones. The best regression models for the quantification of robusta coffee were GBR (pRMSEP of 13.70% and R2 of 0.839) derived based on PCA-extracted features and MLR (pRMSEP of 1.11% and R2 of 0.998) derived based on LDA-extracted features. For the chicory quantification, the same models derived under the same settings as mentioned above also presented the best performances (GBR, pRMSEP = 9.37%, R2 = 0.924; MLR, pRMSEP = 1.54%, R2 = 0.997). The PLSR prediction errors for the quantification of arabica coffee and chicory were 9.90% and 8.09%, respectively. For the classification methods, the LDA model performed well compared to KNN. Generally, some models proved to be effective in quantifying robusta and chicory in roasted arabica coffee. The results of this study indicate that NIR spectroscopy could be a promising method in the coffee industry and other legal sectors for routine applications involving quality control of coffee.
Roasted ground coffees are targets of concern regarding intentional adulteration with cheaper foreign materials because, in this form, it may be difficult to detect due to the small particle size and the dark color. Therefore, a significant interest is developing fast, sensitive, and accurate methodologies to quantify adulterants in roasted coffees. This study investigated the potential of using near-infrared spectroscopy (NIR) to quantity robusta coffee and chicory in roasted arabica coffee. The adulterated arabica coffee samples were composed of robusta coffee or chicory ranging from 2.5 to 30% in increments of 2.5%. Four regression approaches were applied: gradient boosting regression (GBR), multiple linear regression (MLR), k-nearest neighbor regression (KNNR), and partial least squares regression (PLSR). The first three regression models were performed on the features extracted from linear discriminant analysis (LDA) or principal component analysis (PCA). Additionally, two classification methods were applied (LDA and KNN). The regression models derived based on LDA-extracted features presented better performances than PCA ones. The best regression models for the quantification of robusta coffee were GBR (pRMSEP of 13.70% and R2 of 0.839) derived based on PCA-extracted features and MLR (pRMSEP of 1.11% and R2 of 0.998) derived based on LDA-extracted features. For the chicory quantification, the same models derived under the same settings as mentioned above also presented the best performances (GBR, pRMSEP = 9.37%, R2 = 0.924; MLR, pRMSEP = 1.54%, R2 = 0.997). The PLSR prediction errors for the quantification of arabica coffee and chicory were 9.90% and 8.09%, respectively. For the classification methods, the LDA model performed well compared to KNN. Generally, some models proved to be effective in quantifying robusta and chicory in roasted arabica coffee. The results of this study indicate that NIR spectroscopy could be a promising method in the coffee industry and other legal sectors for routine applications involving quality control of coffee.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.