Analytical study for Price Prediction of Bitcoin using Machine Learning and Deep Learning

Abstract

Referring to the recent price boom and bust of cryptocurrencies, Bitcoin is the world's most well-known Cryp-tocurrency, making it enticing to financial market participants. The significant volatility of the Bitcoin conversion scale makes it difficult to predict. As a result, forecasting its behavior is critical for monetary business sectors. However, machine learning based solutions have been proved to be more promising for providing much accurate Bitcoin price predictions. Thus, the aim of this work is to explore and critically analyze the various machine learning based approaches. Further, this work is focused to provide all essential details from the fundamental knowledge of crypto currency particularly Bitcoin to its architecture and price inflation with various factors which is useful for proceeding research and study in the field. The work covers analysis of different supervised machine learning methods assessed to acquire the most pertinent traits for the prediction. The prediction results are also used as inputs to enhance price direction predictions. The outcomes demonstrated that the chosen features and the efficient machine learning and deep learning technique improve accuracy.