Principal component analysis coffee

Bicchi et al. (1995) envisaged the possibility of characterizing green and roasted coffee by the CGA fraction, using the results of HPLC with UV detection and principal component analysis (PCA), and comparing them with the results from sensory evaluation. A direct relationship is not yet fully demonstrated. 

Schlich et al. (1987) proposed a new approach to selecting variables in principal component analysis (PCA) and getting correlations between sensory and instrumental data. Among other studies, Wada et al. (1987a,b) evaluated 39 trade varieties of coffee by coupling gas chromatographic data with two kinds of multivariate analysis. The objective classification was compared with the sensory data (cup test), directly or after statistical treatment. The results were concordant. Murota (1993) used qualitative sensory data to interpret further the results of GC data and canonical discriminant analysis. He could thus suggest which were the components responsible for the flavor characteristics in different coffee cultivars. 

White (1995), not for a sensory analysis but mainly with a view to determining coffee adulterations, used the data of combined headspace GC and high-performance LC for multivariate analysis. Principal component analysis visualized the relationship between samples, and the outlying samples could be identified. The method could be an additional tool for classification and quality control of coffee products. 

Bicchi C.P., Binello A.E., Legovich M.M., Pellegrino G.M. and Vanni A.C. (1993) Characterization of roasted coffee by S-HSGC and HPLC-UV and principal component analysis. J. Agric. Food Chem. 41, 2324-28. 

Shimoda M., Wada K, Shibata K. and Osajima Y. (1985) Studies on aroma of coffee. VII. Evaluation of coffee aroma by principal component analysis. Nippon Shokuhin Kogyo Gakkaishi (J. Jpn. Soc. Food Sci. Technol) 32, 377-85. 

Mendonca et al. have used an electrospray ionization mass spectrometry (ESI-MS) method to identify the CGA profile, which allowed the discrimination of green Arabica and Robusta coffee beans [22]. This method also allowed discrimination between defective and nondefective coffee beans (ESI-MS positive mode). For this kind of identification and discrimination, they used principal component analysis and hierarchical cluster analysis [22]. Alonso-Salces et al. also used a linear discriminant analysis and a partial least-squares discriminant analysis based on HPLC and UV spectra of phenolic (CGAs) and methykanthine contents for a number of green Robusta and Arabica coffee beans from different geographical origins [9]. 

Principal components analysis followed by linear discriminant analysis of the NMR spectra from 98 instant spray dried coffees, obtained from 3 different producers, correctly attributed 99% of the samples to their manufacturer. Blind testing of the PCA model with a further 36 samples of instant coffee resulted in a 100% success rate in identifying the samples from the 3 manufacturers. Coffees from one manufacturer were also assigned into 2 groups using these techniques... 

X Yang, T Peppard. Solid-phase microextraction for flavor analysis. J Agric Food Chem 42 1925 1930, 1994 CP Bicchi, OM Panero, GM Pellegrino, AC Vanni. Characterization of roasted coffee and coffee beverages by solid phase microextraction-gas chromatography and principal component analysis. J Agric Food Chem 45 4680-4686, 1997. 

Seven types of espresso coffee were classified by Pardo and Sberveglieri with a system composed of an electronic nose and an SVM with polynomial and Gaussian RBF kernels.For each coffee type, 36 measurements were performed with an electronic nose equipped with five thin-film semiconductor sensors based on SnOi and Ti-Fe. The output signal from sensors was submitted to a PGA analysis whose principal components (between 2 and 5) represented the input data for the SVM classifier. The error surface corresponding to various kernel parameters and number of input principal components was investigated. 

---