Research Article
Prediction of Defective Software Modules Using Class Imbalance Learning
Table 6
Comparison on the basis of
-measure values on eight datasets.
| Dataset | SVM | CBNN | NB | RF | LR | -NN | BBN | C4.5 | LSTSVM | WLSTSVM |
| CM1 | 0.1900 | 0.4664 | 0.4212 | 0.3398 | 0.3100 | 0.4014 | 0.4873 | 0.2975 | 0.5430 | 0.8191 | KC1 | 0.2749 | 0.5908 | 0.3757 | 0.3637 | 0.4900 | 0.5341 | 0.4003 | 0.3512 | 0.8522 | 0.7639 | PC1 | 0.6033 | 0.5172 | 0.4114 | 0.4365 | 0.3356 | 0.3606 | 0.6018 | 0.4102 | 0.6884 | 0.8019 | PC3 | 0.6200 | 0.6140 | 0.3391 | 0.2828 | 0.3315 | 0.3534 | 0.3929 | 0.4395 | 0.5186 | 0.6371 | PC4 | 0.6048 | 0.6542 | 0.4582 | 0.5823 | 0.5078 | 0.5026 | 0.5527 | 0.4976 | 0.6252 | 0.8236 | MC2 | 0.5411 | 0.7793 | 0.4439 | 0.5952 | 0.4130 | 0.4237 | 0.4643 | 0.5464 | 0.7568 | 0.8318 | KC2 | 0.6362 | 0.7269 | 0.5285 | 0.6018 | 0.6485 | 0.6516 | 0.6809 | 0.6022 | 0.8441 | 0.8478 | KC3 | 0.3667 | 0.5692 | 0.4899 | 0.4859 | 0.3680 | 0.5597 | 0.6164 | 0.3443 | 0.6318 | 0.6762 |
|
|