A Framework for Prediction of Response to HCV Therapy Using Different Data Mining Techniques

Table 4

Performances of 6 iterations of AC with selected features.

AC number	Generated rules	Filtered rules	TP	TN	Positive predictive value %	Negative predictive value %	Sensitivity %	Specificity %	AUC %	Accuracy %
AC1	116	56	10	25	76.9	67.6	45.5	89.3	67.4	70
AC2	119	58	9	29	81.8	74.4	47.4	93.5	70.5	76
AC3	113	60	11	29	78.6	80.5	61.1	90.6	75.9	80
AC4	112	59	12	29	85.7	80.6	63.2	93.5	78.4	82
AC5	118	60	13	31	86.7	88.6	76.5	93.9	85.2	88
AC6	119	61	18	28	100	87.5	81.8	100	90.9	92
Average			12	29	84.95	79.87	62.58	93.47	78.05	81.33