Learning to Discriminate Adversarial Examples by Sensitivity Inconsistency in IoHT Systems

Table 2

Detection performance of different machine learning model architectures. Bold values indicate the optimal results.

Dataset	Machine learning model	Accuracy (%)	Recall (%)	F1-score (%)
IMDB	Random forest	95.7	96.0	95.1
	XGBoost	96.1	97.2	96.4
	LightGBM	96.3	94.1	95.8
	SVM	92.3	93.2	92.1
	AdaBoost	95.2	96.0	94.9
AG’s news	Random forest	89.5	88.1	89.0
	XGBoost	92.7	91.9	92.2
	LightGBM	91.7	80.4	91.3
	SVM	90.4	90.0	89.1
	AdaBoost	88.8	88.9	88.8