Machine Learning Approaches to Predict Patient’s Length of Stay in Emergency Department

Table 6

Confusion matrix for 5-CAT for each algorithm.

Algorithm	Confusion matrix
Logistic	a	b	c	d	e
	3	15	11	3	0
	16	56	60	6	0
	7	35	168	2	0
	2	2	2	4	1
	0	0	1	3	3
Naive Bayes	a	b	c	d	e
	2	16	11	2	1
	7	68	63	0	0
	1	24	187	0	0
	3	4	0	3	1
	0	1	2	3	1
SMO	a	b	c	d	e
	4	15	12	1	0
	10	62	65	1	0
	1	24	187	0	0
	3	4	0	1	3
	0	2	1	1	3
lazy.IBk	a	b	c	d	e
	0	17	15	0	0
	0	59	79	0	0
	0	13	199	0	0
	0	10	1	0	0
	0	4	3	0	0
Decision stump	a	b	c	d	e
	0	17	15	0	0
	0	59	79	0	0
	0	13	199	0	0
	0	10	1	0	0
	0	4	3	0	0
REP tree	a	b	c	d	e
	0	17	15	0	0
	0	61	77	0	0
	0	21	191	0	0
	0	10	1	0	0
	0	4	3	0	0
Random Forest	a	b	c	d	e
	0	19	13	0	0
	3	57	78	0	0
	0	43	169	0	0
	0	9	2	0	0
	0	4	3	0	0
MLP	a	b	c	d	e
	0	0	32	0	0
	0	0	138	0	0
	0	0	212	0	0
	0	0	11	0	0
	0	0	7	0	0