A Feature-Driven Decision Support System for Heart Failure Prediction Based on Statistical Model and Gaussian Naive Bayes

Table 3

Results of different subsets of features for the heart disease dataset.

n			Spec. (%)	Sens. (%)	MCC
1	78.88	75.36	81.63	75.60	0.573
2	88.11	78.26	79.59	82.92	0.622
3	84.44	81.64	89.79	78.04	0.686
4	86.66	79.71	91.83	80.48	0.732
5	86.66	80.19	93.87	78.04	0.734
6	86.66	80.67	89.79	82.92	0.730
7	90.00	82.12	91.83	87.80	0.798
8	90.00	83.57	93.87	85.36	0.799
9	93.33	84.05	97.95	87.80	0.868
10	90.00	81.64	93.87	85.36	0.799
11	90.00	82.60	93.87	85.36	0.599
12	90.00	84.05	93.87	85.36	0.799
13	90.00	82.12	93.87	85.36	0.799