A Hybrid Approach by CEEMDAN-Improved PSO-LSTM Model for Network Traffic Prediction

Table 1

Errors of LSTM prediction models under different parameter combinations.

Model	Parameters	RMSE	MAE	MAPE (%)
LSTM101	(look_back, LN) = (1,1)	71.82	41.68	32.17
LSTM105	(look_back, LN) = (5, 1)	66.81	40.39	32.77
LSTM110	(look_back, LN) = (10,1)	67.08	41.58	34.11
LSTM201	(look_back, LN) = (1,2)	71.65	42.28	33.29
LSTM205	(look_back, LN) = (5,2)	69.58	40.57	31.19
LSTM210	(look_back, LN) = (10,2)	70.43	43.01	34.97