A Medical Decision Support System to Assess Risk Factors for Gastric Cancer Based on Fuzzy Cognitive Map

Table 4

Performance metrics.

Classifiers	+	High	Medium	Low	Class recall	Class precision	Overall accuracy	RMSE	MAE
Decision trees	High	30	10	1	53.57	73.17	76.78	0.5120	0.721
	Medium	16	52	0	81.25	76.47
	Low	10	2	47	97.91	79.66
Naïve Bayes	High	40	8	5	71.42	75.47	80.35	0.334	0.645
	Medium	8	56	4	87.5	77.77
	Low	8	0	39	81.25	82.97
SVM	High	46	2	4	82.14	88.46	86.9	0.193	0.342
	Medium	0	60	4	93.75	93.75
	Low	10	2	40	83.3	76.92
MLP-ANN	High	49	2	7	87.5	84.48	90.47	0.248	0.097
	Medium	4	58	4	90.62	87.87
	Low	3	4	45	93.75	86.53
Proposed model	High	55	1	1	98.21	96.49	95.83	0.173	0.0471
	Medium	1	60	1	93.75	96.77
	Low	0	3	46	95.83	93.87