Attention Mechanism with Spatial-Temporal Joint Deep Learning Model for the Forecasting of Short-Term Passenger Flow Distribution at the Railway Station

Table 4

Comparison of GCN and DeepWalk-based method prediction errors for different parameters.

	Method	RMSE
1	ST-Bi-LSTM	6.605
2	GCN^2-Bi-LSTM (16)	6.755
3	GCN^2-Bi-LSTM (32)	6.617
4	GCN^2-Bi-LSTM (64)	7.061
5	GCN^2-Bi-LSTM (96)	6.645
6	GCN^2-Bi-LSTM (128)	7.033
7	GCN^3-Bi-LSTM (16)	6.477
8	GCN^3-Bi-LSTM (32)	6.420
9	GCN^3-Bi-LSTM (64)	7.007
10	GCN^3-Bi-LSTM (96)	7.020
11	GCN^3-Bi-LSTM (128)	6.627