AGG: A Novel Intelligent Network Traffic Prediction Method Based on Joint Attention and GCN-GRU

Table 4

Comparison results between AGG model with other baseline models.

T (min)	Metric	HA	ARIMA	SVR	GRU	GCN-GRU	AGG
10	RMSE	6.1774	5.6241	5.5817	5.4932	5.4761	5.2721
	MAE	3.2447	3.9729	2.8332	2.7525	2.7009	2.5788
	ACC	0.6785	0.6264	0.7095	0.7114	0.7150	0.7256
		0.7613	0.0356	0.8051	0.8078	0.8124	0.8261
	EVS	0.7613	0.0008	0.8065	0.8086	0.8125	0.8265
20	RMSE	6.1774	5.8655	5.8321	5.8211	5.8189	5.7209
	MAE	3.2447	3.9767	2.9926	2.9836	2.9732	2.8362
	ACC	0.6785	0.6262	0.6856	0.6898	0.6973	0.7024
		0.7613	0.0356	0.7749	0.7853	0.7888	0.7959
	EVS	0.7613	0.0009	0.7896	0.7914	0.7931	0.7969