Institutional Repository
Technical University of Crete
Technical University of Crete
EN
|
EL
| URI: | http://purl.tuc.gr/dl/dias/C875E5D0-1A2F-40FC-AE6B-322A65CDB41A | ||
| Year | 2025 | ||
| Type of Item | Diploma Work | ||
| License |
|
||
| Bibliographic Citation | Vasileios Petrakis, "Fuzzy techniques for cryptocurrency forecasting", Diploma Work, School of Production Engineering and Management, Technical University of Crete, Chania, Greece, 2025 https://doi.org/10.26233/heallink.tuc.102292 | ||
| Appears in Collections |
In this thesis, the prediction of cryptocurrency closing prices is pursued through the application of an adaptive neuro-fuzzy inference system (ANFIS) and the implementation of fuzzy sets (fuzzy type-2). Cryptocurrencies, while they do not represent traditional currencies (fiat currencies) or investment assets in a conventional way, have nonetheless attracted the interest of many investors, resulting in the circulation of vast amounts of capital in the global economy due to cryptocurrencies. Their price fluctuations provide an opportunity for many professional cryptocurrency traders to earn significant profits, while inexperienced traders are subject to substantial losses. The ANFIS system and the fuzzy type-2 method were selected from a variety of forecasting methods due to the advantages offered by the combination of fuzzy logic and artificial neural networks, which form the basis of the ANFIS algorithm. In the fuzzy type-2 system, model optimization will be performed using the particle swarm algorithm (PSO). The results will be compared with traditional forecasting methods, such as auto regression (AR) and auto regression moving average (ARMA), for further evaluation of the outcomes.
Copyright © DIAS 2013
