Research Article
A Low Computational Cost Method for Mobile Malware Detection Using Transfer Learning and Familial Classification Using Topic Modelling
Table 11
Performance outcomes for distinct topics quantity.
| Data set | Core topics quantity | F-Score | Precision | Recall |
| Drebin | 10 | 0.977 | 0.987 | 0.969 | 25 | 0.950 | 0.956 | 0.945 | 45 | 0.880 | 0.947 | 0.823 |
| Malgenome | 10 | 0.880 | 0.981 | 0.798 | 25 | 0.971 | 0.988 | 0.955 | 45 | 0.947 | 0.988 | 0.910 |
| AAGM | 10 | 0.878 | 0.948 | 0.818 | 25 | 0.956 | 0.961 | 0.952 | 45 | 0.966 | 0.988 | 0.946 |
|
|