Table 4 Performance comparison.

Methods	Features	Accuracy	Precision	Recall	F1 score
SVM	Reference	0.68	0.84	0.68	0.74
	SDC	0.75	0.83	0.75	0.78
	SDC-L	0.81	0.93	0.81	0.85
k-NN	Reference	0.90	0.85	0.90	0.87
	SDC	0.90	0.85	0.90	0.87
	SDC-L	0.88	0.84	0.88	0.85
RF	Reference	0.87	0.81	0.87	0.88
	SDC	0.88	0.83	0.88	0.84
	SDC-L	0.88	0.80	0.88	0.84
MLP	Reference	0.89	0.87	0.89	0.88
	SDC	0.89	0.87	0.89	0.88
	SDC-L	0.95	0.95	0.95	0.95
cNN	Reference	0.89	0.82	0.89	0.85
	SDC	0.85	0.75	0.85	0.80
	SDC-L	0.88	0.82	0.88	0.84
cNN + LSTM	Reference	0.91	0.92	0.91	0.88
	SDC	0.90	0.86	0.90	0.88
	SDC-L	0.94	0.94	0.94	0.93

SVM, support vector machine; k-NN, k-nearest neighbors; RF, random forest; MLP, multi-layer perceptron; cNN, convolutionary neural network, LSTM, long short term memory.

Quick links

Search