TN-GTN: fault diagnosis of aircraft wiring network over edge computing

EURASIP Journal on Wireless Communications and Networking

Table 5 Experiment results in the fault classification task

		Fault1			Fault2			Fault3			Fault4
		Precision	Recall	F1-score	Precision	Recall	F1-score	Precision	Recall	F1-score	Precision	Recall	F1-score
20% training	ANN	0.92	0.76	0.83	0.61	0.74	0.67	0.79	0.73	0.76	0.62	0.62	0.62
20% training	TN-GTN	0.96	0.85	0.91	0.77	0.87	0.82	0.9	0.85	0.88	0.75	0.74	0.76
40% training	ANN	0.95	0.78	0.86	0.6	0.78	0.68	0.89	0.78	0.83	0.67	0.75	0.73
40% training	TN-GTN	0.96	0.88	0.92	0.77	0.87	0.82	0.95	0.9	0.92	0.73	0.75	0.77
60% training	ANN	0.97	0.82	0.89	0.62	0.78	0.69	0.94	0.8	0.87	0.67	0.75	0.71
60% training	TN-GTN	0.99	0.91	0.95	0.84	0.91	0.87	0.95	0.9	0.92	0.86	0.88	0.89
80% training	ANN	0.99	0.82	0.9	0.61	0.78	0.76	0.93	0.98	0.95	0.8	0.8	0.8
80% training	TN-GTN	0.99	0.94	0.96	0.89	0.91	0.88	0.98	0.96	0.95	0.9	0.91	0.91