From: Android malware category detection using a novel feature vector-based machine learning model
Existing methods | Accuracy | Precision | Recall | F1-score | Method used |
---|---|---|---|---|---|
Martin et al. (2018) | 0.761 | 0.755 | 0.76 | 0.755 | CNN + Markov chains |
Martin et al. (2018) | 0.818 | 0.807 | 0.818 | 0.802 | RF + Markov chains |
Arindaam Roy. et al. (2020) | 0.887 | 0.895 | 0.819 | 0.855 | SVM + Feature aggregation |
Nicheporuk et al. (2020) | 0.933 | 0.938 | 0.937 | 0.938 | CNN + word2vec technology-based Feature vectorization |
Samaneh et al. (2022) | 0.982 | 0.982 | 0.982 | 0.982 | Semi-supervised DNNs |
Hashem A. El Fiky et al. (2021) | 0.9689 | Not available | 0.6646 | Not available | RF |
Proposed Method | 0.9870 | 0.987 | 0.987 | 0.987 | RF + Huffman encoding-based Feature Vector Generation |