Prediction of Telecommunication Network Fraud Crime Based on Regression-LSTM Model

Table 7

Comparison of MAE, MAPE, RMSE, ACC, and CPU time of different models.

Models	MAE	MAPE	RMSE	ACC (%)	CPU time (second)
ARIMA	0.141	17.330	0.161	80.21	1.16
LSTM	0.130	16.082	0.171	78.98	1.82
Regression	0.124	15.055	0.157	81.89	0.07
Additive-ARIMA-LSTM	0.124	15.652	0.148	81.68	3.01
Multiplicative-ARIMA-LSTM	0.118	14.968	0.155	81.07	3.77
Regression-LSTM	0.122	16.751	0.149	86.80	1.92