A Novel Method for Sea Surface Temperature Prediction Based on Deep Learning

Table 3

The results of p4 on the Yellow Sea dataset which predict 1 day’s SST value (H = 7).

Methods	RMSE	RMSPE	MAPE	MAE (°C)	ACC
SVR	1.1780	6.4541	3.4226	0.5137	0.9482
SVM	1.4618	10.0451	6.2955	0.8457	0.9360
ARIMA	1.4307	9.1634	5.8763	0.8147	0.9398
BPNN	3.6241	14.1903	13.0661	3.7138	0.8503
RBFNN	2.8173	14.1829	12.1397	2.5015	0.8711
RNN	3.0071	15.2845	13.7467	2.4630	0.8548
GRU	1.0814	9.9060	7.8468	1.8768	0.9287
LSTM	1.8271	10.0376	8.8478	2.3915	0.9194
Updated-LSTM	0.7930	2.1523	2.4121	0.8415	0.9613
GRU-SVM	1.5029	7.3219	7.0121	1.5015	0.9259
WNN	0.9180	2.6591	2.9758	1.2371	0.9528
CEEMDAN-LSTM	1.8274	13.879	11.2769	2.0746	0.9017
DGCnetwork	0.3637	1.7382	1.2915	0.2673	0.9830