Kernel PCA feature extraction and the SVM classification algorithm for multiple-status, through-wall, human being detection

EURASIP Journal on Wireless Communications and Networking

Table 2 Experimental results compared with PCA + SVM, KPCA + SVM

Classification algorithm parameters	PCA + SVM		KPCA + SVM
Correctly classified instances	633	79.125%	654	81.75%
Incorrectly classified instances	167	20.875	146	18.25%
Kappa statistic	0.7217		0.7567
Mean absolute error	0.2827		0.2794
Root mean squared error	0.3604		0.3558
Relative absolute error	75.3889%		74.5%
Root relative squared error	83.2388%		82.1584%
Total number of instances	800		800

KPCA kernel principal component analysis, PCA principal component analysis, SVM support vector machine