Stocks Forecast in the Brazilian Petrochemical Sector with the Use of Machine Learning

How to contemplate several types of data and generate an efficient analysis? It was possible to notice an opportunity to use Machine Learning (ML) with artificial intelligence. This study uses ML techniques and automated data extraction to focus on forecast price trends in assets of the petrochemical sector in the Brazilian market. This work aims to analyze the efficiency of forecasting asset trends using ML algorithms: Random Forest, Support Vector Machine, and Artificial Neural Networks. For training the models, we use technical indicators. We measured the results through performance analysis, with indicators such as recall, precision, accuracy, and specificity. As the main findings, it was possible to verify the efficiency of using ML techniques to forecast price trends of assets in the financial market, with an accuracy that varied between 82.4% and 96.4% - being the best algorithm for trend prediction for these proposed assets, the Random Forest