COVID-19 in Iran: Forecasting Pandemic Using Deep Learning
Table 2
Comparison of performance in test stage on nine selected countries.
Model
BI
DI
Lag
MAPE
score
RMSE
NRMSE
RF
2.34
0.99119
19497
0.069109
2.6
0.98764
23103
0.081891
2.7
0.98741
23308
0.082617
MLP
1.32
0.99422
499
0.001769
1.7
0.99348
530
0.001879
1.54
0.99380
517
0.001833
LSTM-R
2.260
0.98088
3242.91
0.011495
2.032
0.98185
1065.41
0.003776
2.025
0.99023
1032.34
0.003659
LSTM-E
1.213
0.99521
750.60
0.002661
1.251
0.99671
610.86
0.002165
0.911
0.99832
580.41
0.002057
M-LSTM
0.726
0.99884
530.74
0.001881
0.550
0.99981
490.17
0.001737
0.509
0.99997
458.12
0.001624
represents only confirmed cases as input and shows the confirmed cases along with death and recovered cases. BI: basic information; DI: detailed information. The performance is evaluated on confirmed cases as output.
