Implementations of Deep Learning in Cryptocurrency Based on Comparisons of Various Scenarios

DOI

10.2991/978-94-6463-821-9_77How to use a DOI?

Keywords

Cryptocurrency prediction; CNN; LSTM; GRU; self-attention; Transformer

Abstract

As a matter of fact, the cryptocurrency market attracts plenty of investors on account of its ultra-high volatility. To realize accurate prediction of the price for cryptocurrency, lots of state-of-art machine learning scenarios are applied. With this in mind, this study explores deep learning’s role in Bitcoin price prediction, focusing on LSTM, GRU, CNN and transformer models. These models excel at identifying long-term patterns and nonlinear features in time series data. In reality, bitcoin’s volatility makes price prediction challenging, as traditional methods fall short. According to the analysis, LSTM and GRU address gradient issues in RNNs, improving accuracy through specialized gating mechanisms. Hybrid models like DL-GuesS, LSTM with Self-Attention, IFA-BiLSTM, and MFB are also examined. Transformer Frameworks are applied for prediction and achieve good results. These integrate multi-source data (e.g., historical prices, social media sentiment, news) to enhance prediction accuracy. The MFB model, combining BiLSTM and BiGRU with BorutaShap for feature selection, represents a cutting-edge approach. Overall, these results shed light on guiding further exploration of cryptocurrency price prediction.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Zhanyu Li
AU  - Siwei Mao
PY  - 2025
DA  - 2025/08/31
TI  - Implementations of Deep Learning in Cryptocurrency Based on Comparisons of Various Scenarios
BT  - Proceedings of the 2025 2nd International Conference on Mechanics, Electronics Engineering and Automation (ICMEEA 2025)
PB  - Atlantis Press
SP  - 798
EP  - 807
SN  - 2352-5401
UR  - https://doi.org/10.2991/978-94-6463-821-9_77
DO  - 10.2991/978-94-6463-821-9_77
ID  - Li2025
ER  -