• Gonzalo Lara de Leyva
4. semester, Master of Science (MSc) in Cyber Security (Master Programme)
In the last decade, cryptocurrencies have become ever more popular but so have cryptocurrency investment scams. In order for investors to make safe, well-informed investments in the cryptocurrency market, this project proposes a price forecasting tool, named CincoCrypto, which is developed to be a simple, open-source alternative targeted to all audiences. Four Machine Learning models are trained in the tool, using financial data from 1-year datasets to make 1-day ahead predictions. The results show that the Space Vector Regression model is the fastest whereas the Convolutional Neural Network model is the most accurate of the four.
Publication date21 May 2023
Number of pages75
ID: 532392440