Table 4.

Classification accuracy of k-NN (k = 1, 3, 10, 20) and SVM (C = 10${}^{-1}$, 10${}^{-2}$, 10${}^{-3}$, 10${}^{-4}$) using ALL feature vector based on IMIM.

	k-NN				SVM
	k = 1	k = 3	k = 5	k = 10	C = 10${}^{-1}$	C = 10${}^{-2}$	C = 10${}^{-3}$	C = 10${}^{-4}$
Accuracy (%)	87.9	88.1	87.5	87.2	88.4	89.2	90.6	88.8
Standard Deviation (%)	8.5	7.9	8.7	8.3	9.4	9.6	8.1	7.8