The Gradient Boosting is a machine learning approach that is widely used due to its high performance and accuracy. The aim of this thesis is find out how good is the performance of Gradient Boosting applied to the price forecasting of Cryptocurrencies and then to flat currencies. The thesis is developed in three sections, the first is an overview of the Cryptocurrencies 's world, the second is an explanation of how Decision trees works and a mayor focus on Gradient Boosting. The last section is the practical part, where there is the application of Gradient Boosting to the price forecasting of cryptocurrencies and then the application of the same algorithm to flat currencies. The aim is to find out if the performance of Gradient Boosting is better for cryptocurrencies forecasting or flat currencies.

Prediction of Cryptocurrency prices using Gradient Boosting machine.

Moreni, Matilde
2020/2021

Abstract

The Gradient Boosting is a machine learning approach that is widely used due to its high performance and accuracy. The aim of this thesis is find out how good is the performance of Gradient Boosting applied to the price forecasting of Cryptocurrencies and then to flat currencies. The thesis is developed in three sections, the first is an overview of the Cryptocurrencies 's world, the second is an explanation of how Decision trees works and a mayor focus on Gradient Boosting. The last section is the practical part, where there is the application of Gradient Boosting to the price forecasting of cryptocurrencies and then the application of the same algorithm to flat currencies. The aim is to find out if the performance of Gradient Boosting is better for cryptocurrencies forecasting or flat currencies.
2020-07-27
File in questo prodotto:
File Dimensione Formato  
877628-1250069.pdf

non disponibili

Tipologia: Altro materiale allegato
Dimensione 4.05 MB
Formato Adobe PDF
4.05 MB Adobe PDF   Richiedi una copia

I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.14247/5605