Predicting cryptocurrency prices with ML-DL models: A hybrid expert system approach

Khaushbakht Kamal, Kainat Mustafa, Rashid Kamal, Yasir Riaz, Chris Nugent, Fouzia Jumani, Sheraz Aslam, Nadeem Javaid

Abstract: Highlights•Hybrid ML-DL model predicts crypto prices using market and sentiment data.•Combines ML for features and DL for time-series to boost accuracy.•Achieves 18.3 % lower RMSE and 6.7 % higher directional accuracy.•The model delivers better Sharpe Ratios.•Enhances decision-making for investors, institutions, and regulators.

External IDs:doi:10.1016/j.eswa.2025.130114