Explainable Artificial Intelligence-Based IoT Device Malware Detection Mechanism Using Image Visualization and Fine-Tuned CNN-Based Transfer Learning Model

Table 3

Performance of deep features with different machine learning classifiers.

Dataset 1
Classifier	Precision	Recall	F1-score	Accuracy
Decision tree	0.98	0.98	0.98	0.982
Logistic regression	0.9	0.9	0.9	0.896
Random forest	0.99	0.99	0.99	0.985
K-nearest neighbor	0.98	0.98	0.98	0.984
Ensemble learning	0.98	0.98	0.98	0.984
Dataset 2
Decision tree	0.88	0.88	0.88	0.875
Logistic regression	0.25	0.27	0.25	0.273
Random forest	0.91	0.91	0.91	0.91
K-nearest neighbor	0.88	0.88	0.88	0.881
Ensemble learning	0.9	0.9	0.9	0.902

For dataset 1, we found that the proposed method with random forest performed well with an accuracy of 0.985, F1-score of 0.99, precision of 0.99, and recall of 0.99. Similarly, for dataset 2, we found that the proposed method with random forest performed well with an accuracy of 0.91, F1-score of 0.91, precision of 0.91, and recall of 0.91).