Bulletin of Electrical Engineering and Informatics Vol. 14, No. 1, February 2025, pp. 513~523 ISSN: 2302-9285, DOI: 10.11591/eei.v14i1.7377  513 Journal homepage: http://beei.org A new deep learning approach for predicting high-frequency short-term cryptocurrency price Issam Akouaouch 1 , Anas Bouayad 2 1 Applied Physics, Computer Science, and Statistics Laboratory, Department of Informatics, Faculty of Science, Sidi Mohamed Ben Abdellah University, Fez, Morocco 2 Artificial Intelligence, Data Sciences, and Emerging Systems Laboratory, Department of Informatics, Faculty of Science, Sidi Mohamed Ben Abdellah University, Fez, Morocco Article Info ABSTRACT Article history: Received Aug 11, 2023 Revised Aug 30, 2024 Accepted Sep 28, 2024 Cryptocurrencies are known for their volatility and instability, making them an attractive but risky investment for traders, analysts, and researchers. As the allure of Bitcoin (BTC) and other cryptocurrencies continues to grow, so does the interest in predicting their prices. To forecast the market rate and sustainability of cryptocurrencies, this study uses machine learning-based time series analysis. The study employs forecast periods ranging from 1 to 10-minutes to categorize the consistency of the market. High-frequency pricing of cryptocurrencies is anticipated with a timestep of up to 10 seconds using various deep learning (DL) models. A hybrid model combining long short-term memory (LSTM) and gated recurrent unit (GRU) is created and compared with standard LSTM and GRU models. Mean squared error (MSE) is the benchmark for estimating the models' performance. The study achieves better results than benchmark models, with MSE values for BTC, Cardano (ADA), and Cosmos (ATOM) in a 5-minute window size being 0.000192, 0.000414, and 0.000451, respectively, and for a 10-minute window size being 0.000212, 0.000197, and 0.000746. Compared to existing models, the suggested model offers a high price predicting accuracy. This study on crypto price prediction using machine learning applications is a preliminary investigation into the topic. Keywords: Cryptocurrencies Deep learning Forecasting Gated recurrent unit Long short-term memory Time series This is an open access article under the CC BY-SA license. Corresponding Author: Issam Akouaouch Applied Physics, Computer Science, and Statistics Laboratory, Department of Informatics Faculty of Science, Sidi Mohamed Ben Abdellah University Fez, Morocco Email: Issam.akouaouch@usmba.ac.ma 1. INTRODUCTION Cryptocurrencies have emerged as a novel and disruptive financial technology, challenging traditional notions of currency and investment. These digital assets, underpinned by blockchain technology, offer a decentralized alternative to conventional fiat currencies, making them immune to central authority interference or manipulation [1]. The allure of cryptocurrencies, such as bitcoin (BTC) and ethereum (ETH), lies not only in their technological innovation but also in their potential for speculative investment. Due to their inherent volatility, cryptocurrencies present a unique challenge for predictive modeling, attracting significant interest from researchers and investors alike. Despite the growing body of research on cryptocurrency price prediction, there remains a substantial gap in accurately forecasting short-term price movements. The volatile nature of the cryptocurrency market, driven by factors such as market sentiment, regulatory news, and technological advancements, complicates