AIRES, L. G.; http://lattes.cnpq.br/8111670868470553; AIRES, Lucas Gomes.
Resumo:
The cryptocurrency market has gained a lot of visibility over the last few years. According to
Forbes, Brazil is within the top 5 countries with the largest number of investors in the world,
reaching 10 million investors in 2022. Despite the large number of investors, the crypto
market represents a high degree of risk.
This work proposes a predictive method for loss mitigation in the cryptocurrency market
using machine learning techniques. To this end, a tool based on mathematical and graphic
indicators was developed to support decisions to buy and sell the digital asset. The
Gala/USDT currency was chosen for carrying out experiments, the Python language was
used as support for the creation and manipulation of the Data Frame and in the end we
obtained an accuracy of around 90% in the test set that represented 30 % of the Data Frame.