Enhanced fault detection in polymer electrolyte fuel cells via integral analysis and machine learning

Energy Informatics

Table 3 Models' accuracy comparison with sampling for imbalanced data

	Accuracy	ROS Accuracy	SMOTE-OS Accuracy	RUS Accuracy	NearMiss-US Accuracy
LR	0.998	0.998	0.999	0.998	0.988
SVM	0.997	0.998	0.998	0.997	0.059
KNN	0.999	0.999	0.999	0.999	0.998
DT	1	0.999	1	0.999	0.997
RF	1	1	1	0.999	0.999
NB	0.998	0.998	0.998	0.998	0.999
MLP	0.999	0.998	0.997	0.997	0.059

The table presents the results of comparing the accuracy of different models for balanced and unbalanced data in the context of PEFCs. The models evaluated include LR, SVM, KNN, DT, RF, NB, and MLP. The resampling techniques are ROS, SMOTE-OS, RUS, and NearMiss-US