Research Article
N-Gram, Semantic-Based Neural Network for Mobile Malware Network Traffic Detection
Table 4
Description of detecting performance metrics of different domain composition architectures with different text sizes.
| Domain composition type number | Evaluation metrics | Text size (byte) of the transformed data from network flow | 200 | 500 | 700 | 1000 | 1200 | 1500 |
| 1 | Precision | 0.775 | 0.828 | 0.935 | 0.942 | 0.945 | 0.950 | Recall | 0.657 | 0.724 | 0.820 | 0.863 | 0.863 | 0.864 | F1-score | 0.711 | 0.773 | 0.873 | 0.900 | 0.902 | 0.905 | Accuracy | 0.706 | 0.773 | 0.874 | 0.903 | 0.905 | 0.907 |
| 2 | Precision | 0.773 | 0.821 | 0.924 | 0.929 | 0.930 | 0.933 | Recall | 0.657 | 0.721 | 0.810 | 0.850 | 0.849 | 0.854 | F1-score | 0.710 | 0.768 | 0.863 | 0.888 | 0.888 | 0.891 | Accuracy | 0.707 | 0.769 | 0.863 | 0.890 | 0.891 | 0.894 |
| 3 | Precision | 0.778 | 0.818 | 0.918 | 0.920 | 0.924 | 0.926 | Recall | 0.659 | 0.720 | 0.805 | 0.843 | 0.844 | 0.847 | F1-score | 0.714 | 0.765 | 0.858 | 0.880 | 0.883 | 0.885 | Accuracy | 0.709 | 0.766 | 0.858 | 0.883 | 0.886 | 0.887 |
| 4 | Precision | 0.777 | 0.822 | 0.912 | 0.906 | 0.912 | 0.912 | Recall | 0.661 | 0.720 | 0.798 | 0.831 | 0.832 | 0.836 | F1-score | 0.714 | 0.714 | 0.851 | 0.867 | 0.870 | 0.872 | Accuracy | 0.712 | 0.762 | 0.851 | 0.870 | 0.873 | 0.875 |
| 5 | Precision | 0.770 | 0.809 | 0.902 | 0.894 | 0.859 | 0.896 | Recall | 0.660 | 0.712 | 0.791 | 0.820 | 0.821 | 0.823 | F1-score | 0.711 | 0.757 | 0.843 | 0.855 | 0.856 | 0.858 | Accuracy | 0.708 | 0.758 | 0.843 | 0.859 | 0.860 | 0.862 |
| 6 | Precision | 0.784 | 0.836 | 0.900 | 0.946 | 0.944 | 0.949 | Recall | 0.662 | 0.732 | 0.799 | 0.862 | 0.864 | 0.863 | F1-score | 0.718 | 0.781 | 0.847 | 0.902 | 0.902 | 0.904 | Accuracy | 0.713 | 0.781 | 0.848 | 0.904 | 0.905 | 0.906 |
| 7 | Precision | 0.775 | 0.822 | 0.926 | 0.931 | 0.934 | 0.936 | Recall | 0.659 | 0.720 | 0.815 | 0.861 | 0.862 | 0.863 | F1-score | 0.712 | 0.768 | 0.867 | 0.895 | 0.896 | 0.898 | Accuracy | 0.709 | 0.774 | 0.868 | 0.898 | 0.900 | 0.901 |
| 8 | Precision | 0.847 | 0.894 | 0.954 | 0.969 | 0.971 | 0.971 | Recall | 0.715 | 0.784 | 0.848 | 0.880 | 0.881 | 0.881 | F1-score | 0.775 | 0.836 | 0.898 | 0.922 | 0.924 | 0.924 | Accuracy | 0.772 | 0.836 | 0.899 | 0.924 | 0.925 | 0.926 |
| 9 | Precision | 0.848 | 0.896 | 0.954 | 0.969 | 0.971 | 0.971 | Recall | 0.716 | 0.784 | 0.849 | 0.880 | 0.881 | 0.882 | F1-score | 0.776 | 0.836 | 0.899 | 0.923 | 0.924 | 0.924 | Accuracy | 0.772 | 0.837 | 0.899 | 0.924 | 0.925 | 0.926 |
|
|