استخدام شبكة الذاكرة طويلة قصيرة المدى للتنبؤ بأسعار عملة البيتكوين

Abstract

This research paper aimed to predict the daily closing prices of Bitcoin using a developed model of artificial neural networks, represented by a Long-Short Term Memory (LSTM) network, during the period extending from 17/09/2014 - 27/06/2024. The long-short term memory model was built on the Google Colab platform using the Python programming language through a set of libraries belonging to this language. The results of this study showed that there is a high predictive ability for long-short term memory model based on the performance measures adopted bythis researchpaper.