Table 4.

Partial least squares regression (PLSR) models for the quantification of offal using FT-IR spectroscopies.

	model	Na	nb	factors	RMSECc	R 2	RMSECVd	R 2-CV	RMSEPe	R 2-P
FT-IR spectroscopy	quantification of offal	425	180	11	0.07	0.96	0.07	0.96	0.11	0.75
	quantification of beef offal	245	90	9	0.05	0.98	0.06	0.97	0.06	0.97
	quantification of pork offal	245	90	8	0.07	0.96	0.06	0.96	0.14	0.58
	quantification of beef honey comb tripe	75	30	6	0.03	0.99	0.04	0.98	0.05	0.96
	quantification of beef liver	75	30	6	0.03	0.99	0.03	0.99	0.03	0.95
	quantification of beef omasum	75	30	5	0.04	0.99	0.04	0.96	0.03	0.95
	quantification of pork heart	75	30	10	0.04	0.99	0.07	0.96	0.1	0.45
	quantification of pork kidney	75	30	9	0.03	0.99	0.04	0.98	0.07	0.94
	quantification of pork liver	75	30	6	0.03	0.99	0.04	0.99	0.07	0.81
	quantification of beef honey comb tripe & omasum	135	60	5	0.04	0.99	0.04	0.98	0.04	0.95
	quantification of pork heart & kidney	135	60	9	0.05	0.98	0.06	0.97	0.14	0.67

^aN, number of spectra for calibration. ^bn, number of spectra for prediction. ^cRMSEC, root mean squares error of calibration. ^dRMSECV, root mean squares error of cross-validation (9-fold). ^eRMSEP, root mean squares error of prediction.