Table 3 Statistical results of the 3-class classification models for CL_r prediction.

Model	Selected descriptors (n)	Training or Test set	Parameter	CR type	RF^a (Model_CL_r_CR)	SVM^a	ANN^a	PLS^a
Without f_u,p	15	Training	Kappa	—	0.34	0.34	0.32	0.29
		Test	Kappa	—	0.32^b	0.19	0.18	0.22
			Sensitivity	R	0.56	0.56	0.56	0.50
				IM	0.29	0.12	0.41	0.18
				S	0.75	0.69	0.47	0.75
			Balanced Accuracy	R	0.7	0.69	0.68	0.68
				IM	0.58	0.50	0.59	0.54
				S	0.68	0.58	0.52	0.59

^aRF, Random forest; SVM, Support Vector Machine; ANN, artificial neural network; PLS, partial least square.
^bThe highest Kappa shown in the test set.

Quick links

Search