Research Article
Boosting Accuracy of Classical Machine Learning Antispam Classifiers in Real Scenarios by Applying Rough Set Theory
Table 4
-score and balanced
-score rates for different
values.
| | = 0.25 | = 0.5 | = 1 |
| NB (+RS) | 0.81 (+0.165) | 0.81 (+0.130) | 0.82 (+0.063) | FB (+RS) | 0.94 (+0.010) | 0.86 (+0.001) | 0.72 (−0.009) | AB (+RS) | 0.92 (+0.057) | 0.91 (+0.041) | 0.88 (+0.012) | SVM (+RS) | 0.95 (+0.031) | 0.93 (+0.018) | 0.88 (−0.004) |
|
|