Análise preditiva com redes neurais artificiais para precificação de grãos no mercado de commodities

Abstract

This paper describes the development of an Artificial Neural Network (ANN) with the objective of forecasting the prices of three agricultural commodities: rice, corn, and soybeans, based on historical data. Scenarios using time series containing price data of the commodities between the years 2004 and 2023 were analyzed and discussed. The developed system allows the automated analysis of future values, using recurrent neural networks and Long Short-Term Memory (LSTM) techniques. The effectiveness of the developed system was evaluated using datasets obtained from public repositories. The results obtained are presented and discussed in detail. The preliminary results of training and prediction demonstrated the effectiveness of the developed computational tool, highlighting the accuracy of the implemented prediction model.