Table 3 Comparative analysis of DCFNN-SOCVDC approach with existing methods39,40,41,42.
Models | \(Accu_{y}\) | \(Prec_{n}\) | \(Reca_{l}\) | \(F1_{score}\) |
|---|---|---|---|---|
SMO classifier | 84.16 | 81.95 | 83.19 | 87.06 |
SVM | 96.72 | 97.22 | 94.65 | 89.74 |
Random forest | 94.25 | 96.39 | 96.30 | 92.29 |
K-nearest | 80.65 | 94.28 | 89.19 | 93.98 |
EDLACNN | 94.10 | 89.30 | 90.28 | 93.28 |
Bagging algorithm | 97.47 | 94.20 | 96.64 | 89.57 |
ACVD-HBOMDL | 98.81 | 97.32 | 95.56 | 97.70 |
AOA method | 98.90 | 98.85 | 98.95 | 98.80 |
DCFNN model | 95.50 | 93.40 | 94.00 | 92.60 |
DCFNN-SOCVDC | 99.05 | 99.05 | 99.05 | 99.04 |
