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

Table 4

, , , and 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), correlation-based feature selection (CFS), Decision Table (DTable), REPTree, and M5P 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 (CFS)	(444, 608, 610, 684)	PSE	0.000	0.000	0.000	0.000	0.000
		P	0.953	0.375	0.752	0.919	0.841
		N	0.927	0.094	0.776	0.878	0.844
		DFD	0.000	0.000	0.000	0.000	0.000
		Weighted AVG	0.759	0.229	0.611	0.677	0.735
DTable	(442, 600, 602, 604, 608, 622)	PSE	0.000	0.000	0.000	0.000	0.000
		P	0.977	0.361	0.764	0.977	0.857
		N	0.878	0.034	0.900	0.878	0.889
		DFD	0.714	0.020	0.625	0.714	0.667
		Weighted AVG	0.791	0.206	0.677	0.791	0.727
REPTree	(442, 452, 460, 568, 604, 606, 626, 654, 666, 698, 1354, 1880)	PSE	0.125	0.022	0.500	0.125	0.200
		P	0.919	0.333	0.767	0.919	0.836
		N	0.878	0.060	0.837	0.878	0.857
		DFD	0.571	0.013	0.667	0.571	0.615
		Weighted AVG	0.772	0.201	0.740	0.772	0.735
DTable (M5P)	(400, 430, 410, 448, 482, 540, 608, 618, 626, 968, 1376, 1710, 1874, 1880, 2476, 2494)	PSE	0.042	0.007	0.500	0.042	0.077
		P	0.942	0.375	0.750	0.942	0.835
		N	0.805	0.060	0.825	0.805	0.815
		DFD	0.571	0.026	0.500	0.571	0.533
		Weighted AVG	0.753	0.222	0.720	0.753	0.701