Epatha, Leono and Aedah, Abd Rahman and Hoga, Saragih (2021) Deep-Recurrent Neural Networks Approach for Indonesian Banks Term Deposit Interest Rates Prediction. In: 5th International Conference on Informatics and Computational Sciences (ICICoS).
Full text not available from this repository.Abstract
This digital era has brought a significant impact on banking. Banks have become an intensive subject of data and must optimize data usage for more insight. Banks can explore their data to increase their productivity and sales. A study in 2019 showed XYZ Bank in Indonesia has not yet explored data and did not optimize it to understand customers and their needs better. This study tried to find the best Recurrent Neural Network (RNN) technique to predict Indonesian banks' term deposit interest rates by comparing three popular RNN variants. Those RNN techniques were Simple RNN, LSTM, and GRU, which used historical data from 22 prominent banks in Indonesia covers 2019–2021. This study found that RNN with Simple RNN technique outperformed LSTM and GRU. Simple RNN brought the smallest mean of RMSE with 1,48% RMSE reduction from LSTM and 1,69% RMSE reduction from GRU
| Item Type: | Conference or Workshop Item (Paper) |
|---|---|
| Uncontrolled Keywords: | Productivity, Recurrent neural networks, Economic indicators, Pandemics, Computer architecture, Banking, Logic gates |
| Depositing User: | Muhamad Aizat Nazmi Mohd Nor Hamin |
| Date Deposited: | 06 Apr 2022 07:29 |
| Last Modified: | 06 Apr 2022 07:29 |
| URI: | http://ur.aeu.edu.my/id/eprint/935 |
Actions (login required)
 |
View Item |
