Table 7: The confusion matrix for the Bayesian model averaging of artificial neural networks. The last but one line lists the total number of light curves (TOT) to define every class. The last line lists the correct classification rate (CC) for every class separately as measured by 10-fold cross validation. Separation between: A: Cepheids; B: long period variables; C: RR Lyrae stars.

A					B			C
	CLCEP	PTCEP	RVTAU	DMCEP		MIRA	SR		RRAB	RRC	RRD
CLCEP	190	17	2	0	MIRA	139	6	RRAB	126	5	1
PTCEP	4	3	3	0	SR	6	36	RRC	3	23	0
RVTAU	1	2	6	0	TOT	145	42	RRD	0	1	56
DMCEP	0	2	2	95	CC(%)	95.86	85.71	TOT	129	29	57
TOT	195	24	13	95				CC(%)	97.67	79.31	98.24
CC(%)	97.43	12.50	46.15	100.00