Machine Learning Applied to Near-Infrared Spectra for Chicken Meat Classification

Table 6

Traditional attributes and poultry classification results (pale, soft, and exudative (PSE); dark, firm, and dry (DFD); normal (N) or pale (P)) obtained by support vector machine (SVM) and REPTree in mean of average true positive (TP), false positive (FP), Precision, Recall, and F-Measure after 30 repetitions.

Algorithm	Attributes	Class	TP	FP	Precision	Recall	F-Measure
SVM	(pH, , , , /, WHC, texture, chroma, hue)	PSE	0.000	0.000	0.000	0.000	0.000
		P	0.965	0.361	0.761	0.965	0.851
		N	0.951	0.085	0.796	0.951	0.867
		DFD	0.000	0.000	0.000	0.000	0.000
		Weighted AVG	0.772	0.219	0.621	0.772	0.688
SVM	(pH, )	PSE	0.000	0.000	0.000	0.000	0.000
		P	0.965	0.347	0.769	0.965	0.856
		N	0.976	0.085	0.8	0.976	0.879
		DFD	0.000	0.000	0.000	0.000	0.000
		Weighted AVG	0.778	0.211	0.626	0.778	0.694
REPTree	(pH, , , , /, WHC, texture, chroma, hue)	PSE	0.958	0.007	0.958	0.958	0.960
		P	0.988	0.028	0.977	0.988	0.968
		N	0.976	0.000	1.000	0.976	0.986
		DFD	1.000	0.000	1.000	1.000	1.000
		Weighted AVG	0.981	0.016	0.981	0.981	0.981
REPTree	(pH, )	PSE	0.958	0.007	0.958	0.958	0.960
		P	0.988	0.028	0.977	0.988	0.968
		N	0.976	0.000	1.000	0.976	0.986
		DFD	1.000	0.000	1.000	1.000	1.000
		Weighted AVG	0.981	0.016	0.981	0.981	0.981