Financial Expenses Forecasting System Based on Recurrent Neural Networks
Sebastian Gonzales-Abad, Hugo Quispe-Chavez, Alvaro Aures-García · 2024
This research is based on a financial expense forecasting system that takes advantage of recurrent neural networks (RNN) to improve spending control among young Peruvians. Furthermore, a comprehensive analysis of the existing literature on various forecasting systems, mainly focused on finance, is presented. This analysis has facilitated the development of a neural network model designed for mobile applications and conceptualized within a RESTful API framework. The Root Mean Square Error (RMSE) of the model developed, under the concept of RNN, has reached 0.410, with a total sample of 40 young people, who stated that the application has adequately provided them with information about the expenses they have made in periods of different times.