Estimating cocoa bean parameters by FT-NIRS and chemometrics analysis. (1st June 2015)

• Record Type: Journal Article
• Authors: Teye, Ernest; Huang, Xingyi; Sam-Amoah, Livingstone K.; Takrama, Jemmy; Boison, Daniel; Botchway, Francis; Kumi, Francis
• Is Part Of: Food chemistry. Volume 176(2015)
• Page Start: 403
• Page End: 410
• Publication Date: 2015-06-01
• Subjects: FT-NIRS -- Cocoa bean categories -- pH -- Fermentation index -- Multivariate algorithms; Food -- Analysis -- Periodicals; Food -- Composition -- Periodicals; 664
• DOI: 10.1016/j.foodchem.2014.12.042
• Languages: English
• ISSNs: 0308-8146

Abstract

Highlights: Smooth-1der PCA gave a satisfactory cluster score plot. The best classification model was derived by Smooth-1der BPANN. Efficient variable selection by synergy interval PLS (SiPLS). Optimal regression model was achieved by Si-BPANNR for pH and fermentation index. The relationship between NIRS and cocoa bean category, pH and FI has been developed. Abstract: Rapid analysis of cocoa beans is an important activity for quality assurance and control investigations. In this study, Fourier transform near infrared spectroscopy (FT-NIRS) and chemometric techniques were attempted to estimate cocoa bean quality categories, pH and fermentation index (FI). The performances of the models were optimised by cross-validation and examined by identification rate (%), correlation coefficient ( R pre ) and root mean square error of prediction (RMSEP) in the prediction set. The optimal identification model by back propagation artificial neural network (BPANN) was 99.73% at 5 principal components. The efficient variable selection model derived by synergy interval back propagation artificial neural network regression (Si-BPANNR) was superior for pH and FI estimation. Si-BPANNR model for pH was R pre = 0.98 and RMSEP = 0.06, while for FI was R pre = 0.98 and RMSEP = 0.05. The results demonstrated that FT-NIRS together with BPANN and Si-BPANNR model could successfully be used for cocoa beans examination.