Research Article
LiDAR: A Light-Weight Deep Learning-Based Malware Classifier for Edge Devices
Table 1
Summary of deep learning-based malware classification approaches.
| Name | Algorithm | Accuracy or F1-score | Features |
| MalDozer [11] | CNN | 96% | API call | DL-Droid [12] | MLP | 99% | Permission, etc. | Droid-Sec [13] | DBN | 97% | Permission, API call, etc. | Kim et al. [14] | DNN | 99% | Permission, component, string, opcode, API | DroidDetector [15] | DBN | 97% | API, permission, etc. | DroidDeep [16] | DBN | 99% | Permission, API call, action, component, etc. | Li et al. [17] | DNN | 97% | Permission, API call, etc. | Ganesh et al. [18] | CNN | 93% | Permission | Nix and Zhang [19] | CNN | 99% | API call |
|
|