Research Article
Combat Mobile Evasive Malware via Skip-Gram-Based Malware Detection
Table 2
Subset of dataset trained for tuning of different hyperparameters of skip-gram, window size, and embedding size.
| | | J48 | Naïve Bayes | Random forest | SVM | | Window size | — | 2 | 3 | 2 | 3 | 2 | 3 | 2 | 3 |
| | Embedding size | 50 | 79.77% | 78.91% | 60.51% | 60.44% | 83.89% | 83.56% | 60.01% | 59.78% | | 300 | 81.93% | 81.50% | 60.81% | 62.37% | 87.08% | 87.14% | 59.61% | 60.04% | | 500 | 82.63% | 82.82% | 61.60% | 65.99% | 86.51% | 88.54% | 59.21% | 57.25% |
|
|
