Hybrid Exponential Smoothing Neural Network untuk Peramalan Data Pengguna Pita Lebar di Indonesia

Abstrak

The country of Indonesia has a huge opportunity to realize the potential for broadband users because the population growth rate is high and internet users are increasing each year. For this reason, acceleration is needed in providing telecommunications services and infrastructure that can increase internet access for users. To find out the estimated number of broadband users in the future, in this study proposed a hybrid exponential smoothing neural network forecasting model. By combining these methods, it is possible to take advantage of the forecasting technique of each while overcoming their drawbacks. This study showed that the RMSE value of hybrid model is 29.86% lower than that of the non hybrid. The MAPE value of hybrid model is 99.93% lower than that of the non hybrid model and the MAE value of hybrid model is 38.52% lower than that of the non hybrid model.