NSGA-III-Based Deep-Learning Model for Biomedical Search Engines

Table 2

Training analysis of the NSGA-III-based CNN model.

Model	Specificity	AUC	Sensitivity	F-measure	Accuracy
TCMSearch [17]	78.691.48	78.561.68	78.391.45	78.671.54	78.491.52
SVM [16]	78.691.39	78.521.42	78.541.45	76.751.33	78.681.32
G-Bean [3]	77.651.30	77.251.49	77.761.16	78.150.78	78.210.72
TTA10 [19]	78.180.79	78.190.87	78.320.66	77.260.54	77.310.82
ViLiP [18]	78.280.71	77.390.58	78.310.48	78.450.56	78.340.35
SOSC [1]	78.451.15	78.461.21	78.560.79	78.390.52	78.330.74
GeoNames [20]	78.690.48	78.680.43	78.460.49	78.760.79	78.540.76
CRRP [21]	78.460.46	78.720.55	77.460.43	78.530.42	78.570.55
ASE [22]	79.320.46	79.130.39	79.210.49	49.300.39	79.290.49
CNN [23]	79.350.52	79.430.52	79.290.54	79.280.43	79.280.53
VDP [23]	79.530.52	81.110.35	80.460.46	79.430.36	79.500.39
Proposed DCNN	83.790.46	83.710.79	83.350.51	83.690.50	83.350.52