EUVIMEDDescoperă EUVIMED
Open access · CC BY 4.0

Enhanced Interpretable Forecasting of Cryptocurrency Prices Using Autoencoder Features and a Hybrid CNN-LSTM Model

Wajeeha Badar, Shabana Ramzan, Ali Raza, Norma Latif Fitriyani, Muhammad Syafrudin, Seung Won Lee

Mathematics · 2025-06-07 · 13 (12) · 1908

DOI: 10.3390/math13121908 ↗

At a glance

Journal article in Mathematics, published online on 2025-06-07. Written by Wajeeha Badar and 5 co-authors. Cited 17 times so far (Crossref); 29 references. The full text is freely available (open access, CC BY 4.0). Funded by: SungKyunKwan University and the BK21 FOUR (Graduate School Innovation) funded by the Ministry of Education (MOE, Korea), National Research Foundation of Korea (NRF)., National Research Foundation (NRF) grants funded by the Ministry of Science and ICT (MSIT) and Ministry of Education (MOE), Republic of Korea.

RezumatDate bibliograficeCiteRelated studies

Abstract (original language)

Predicting the price of Bitcoin is crucial, primarily because of the market’s rapid volatility and non-linear environment. For enhanced prediction of the price of Bitcoin, this research proposed a novel interpretable hybrid technique that combines long short-term memory (LSTM) networks with convolutional neural networks (CNN). Deep variational autoencoders (VAE) are used in the stage of preprocessing to determine noticeable patterns in datasets by learning features from historical Bitcoin price data. The CNN-LSTM model additionally implies Shapley additive explanations (SHAP) to promote interpretability and clarify the role of various features. For better performance, the methodology used data cleaning, preprocessing, and effective machine-learning techniques. The hybrid CNN + LSTM model, in collaboration with VAE, obtains a mean squared Error (MSE) of 0.0002, a mean absolute error (MAE) of 0.008, and an R-squared (R2) of 0.99, based on the experimental results. These results show that the proposed model is a good financial forecast method since it effectively reflects the complex dynamics of primary changes in the price of Bitcoin. The combination of deep learning and explainable artificial intelligence improves predictive accuracy as well as transparency, thus qualifying the model as highly useful for investors and analysts.

Funding

  • SungKyunKwan University and the BK21 FOUR (Graduate School Innovation) funded by the Ministry of Education (MOE, Korea)
  • National Research Foundation of Korea (NRF).
  • National Research Foundation (NRF) grants funded by the Ministry of Science and ICT (MSIT) and Ministry of Education (MOE), Republic of Korea

Cite

APA

Badar, W., Ramzan, S., Raza, A., Fitriyani, N. L., Syafrudin, M., & Lee, S. W. (2025). Enhanced Interpretable Forecasting of Cryptocurrency Prices Using Autoencoder Features and a Hybrid CNN-LSTM Model. Mathematics, 13(12), 1908. https://doi.org/10.3390/math13121908

Vancouver

Badar W, Ramzan S, Raza A, Fitriyani NL, Syafrudin M, Lee SW. Enhanced Interpretable Forecasting of Cryptocurrency Prices Using Autoencoder Features and a Hybrid CNN-LSTM Model. Mathematics. 2025;13(12):1908. doi:10.3390/math13121908

BibTeX
@article{Badar2025,
  author = {Badar, Wajeeha and Ramzan, Shabana and Raza, Ali and Fitriyani, Norma Latif and Syafrudin, Muhammad and Lee, Seung Won},
  title = {Enhanced Interpretable Forecasting of Cryptocurrency Prices Using Autoencoder Features and a Hybrid CNN-LSTM Model},
  journal = {Mathematics},
  year = {2025},
  volume = {13},
  number = {12},
  pages = {1908},
  doi = {10.3390/math13121908}
}
Start for free →