Artificial Intelligence-Based Analysis of Cryptocurrency Prices
DOI:
https://doi.org/10.17762/msea.v72i1.2376Abstract
Due to its increasing popularity and acceptance among merchants, cryptocurrency is becoming more and more significant in the effort to change the financial system. Even though a lot of individuals are investing in cryptocurrencies, the dynamic characteristics, unpredictability, and predictability of cryptocurrencies are still largely unknown, dramatically increasing the danger of the investments. It is important to make an effort to comprehend the variables that affect the construction of values. In this study, we investigate the price dynamics of Bitcoin, Etherum, and Ripple using advanced artificial intelligence frameworks of fully linked Artificial Neural Network (ANN) and Long Short-Term Memory (LSTM) Recurrent Neural Network. We discover that LSTM is more effective at utilising useful information than ANN, which primarily relies on long-term history and short-term dynamics.