Journal of Engineering and Applied Science

Table 4 Performance comparison with other models

From: Integrated encoder-decoder-based wide and deep convolution neural networks strategy for electricity theft arbitration

Methods	Training ratio = 60%					Training ratio = 80%
Methods	AUC	ACC	Precision	Recall	F1 Score	AUC	ACC	Precision	Recall	F1 Score
SVM [19]	0. 716	0.897	87	90	88.47	0.729	0.897	87	90	88.47
KNN [20]	0.640	0.892	86	89	87.47	0.630	0.887	85	89	86.95
MLP [27]	0.797	0.907	89	91	89.99	0.818	0.909	90	91	90.49
CNN [34]	0.880	0.938	93	94	93.49	0.910	0.944	94	94	94.00
LDA [25]	0.664	0.883	86	89	87.47	0.682	0.896	87	90	88.47
PM	0.891	0.956	95	96	95.49	0.956	0.974	97	97	97.00

Back to article page