PREDICTING REVENUE OF SHARIA BANKING TRANSACTIONS USING RNN, LSTM, GRU, DECISION TREE, AND QSPM (CASE STUDY: PT BANK TBV SYARIAH)

The banking business will continue to grow significantly along with the increase in the number of transactions carried out by customers through the channels provided by the bank. The variety of products and features offered by PT Bank TBV Syariah to customers means that resources are not optimal. Hence, the bank's revenue growth target still needs to be achieved. This research aims to predict transactions that can affect bank revenues by using transaction data sources for the period January 2022 to February 2024 and which products and features need to be optimized so that it is hoped that banks can run their business appropriately and according to targets. The methods in this research are the RNN, LSTM, GRU, and Decision Tree methods. To enrich information, this research adds QSPM-based strategy analysis using SWOT that the company previously defined. The expected results are to prove the effectiveness of the model used in predicting PT Bank TBV Syariah transaction data to produce MAE, MSE, and RMSE with the lowest values​​, as well as recommendations that PT Bank TBV Syariah must carry out to increase revenue. This research is expected to provide accurate and effective predictions for projecting PT Bank TBV Syariah transaction data, support strategic decision-making, and produce recommendations for significantly increasing bank income.