Comparison of RNN, LSTM, and GRU Methods on Forecasting Website Visitors

Forecasting is the best way to find out number of website visitors. However, many researchers cannot determine which method used solve problem forecasting Several methods have been in research. One today using deep learning methods. This study discusses visitors one family, namely RNN, LSTM, and GRU The comparison made by these three can be get results field forecasting. two types data: First Time Visits Unique Visits. test was carried with epoch parameters starting from 1 500 at layers 1, 3, 5. first-time visit data unique data. Although tested different data, obtained that smallest MSE value LSTM method. each 0.0125 for 0.0265 contribution this research has succeeded showing performance recurrent network values.