Table 3 Experiments on Microsoft BIG-15 Dataset.
From: Semantic lossless encoded image representation for malware classification
Model | Year | Techniques | Normalized image size | Acc | \(F1_{macro}\) | \(FPR_{macro}\) |
|---|---|---|---|---|---|---|
KNN | – | – | \(512\times 512\) | 0.91 | 0.912 | 0.011 |
DT | – | – | \(512\times 512\) | 0.878 | 0.878 | 0.015 |
XGBoost | – | – | \(512\times 512\) | 0.897 | 0.902 | 0.014 |
AlGarni et al.58 | 2022 | EfficientNet | \(224\times 224\) | 0.926 | 0.924 | 0.01 |
Parihar et al.59 | 2022 | ResNet+Xception+EfficientNet | \(224\times 224\) | 0.937 | 0.934 | 0.010 |
Kumar et al.21 | 2022 | Deep CNN(VGG16 based) | \(224\times 224\) | 0.945 | 0.944 | 0.009 |
Son et al.18 | 2022 | k-NN+SVM+CNN | \(32\times 64\) | 0.936 | 0.933 | 0.08 |
Mallik et al.20 | 2022 | VGG16+BiLSTM | \(224\times 224\) | 0.923 | 0.923 | 0.011 |
Rustam et al.22 | 2023 | VVG16+ResNet | \(224\times 224\) | 0.935 | 0.932 | 0.009 |
Proposed method | – | CNN+SPP+Transformer | No Limit | 0.967 | 0.952 | 0.006 |
