Authors : Tiya Joy; Lijin Benny; Jeremy Varghese; Thomas Philip; Bency Wilson

Volume/Issue : Volume 8 - 2023, Issue 4 - April

Google Scholar : https://bit.ly/3TmGbDi

Scribd : https://bit.ly/3nxbxM8

DOI : https://doi.org/10.5281/zenodo.7898024

Abstract : Cryptocurrency is a type of digital or virtual currency that uses cryptography to secure and verify transactions as well as to control the creation of new units, it uses Blockchain properties for the same. Blockchain is a decentralized digital ledger technology that records transactions securely and transparently. Blockchain technology and cryptocurrency are closely connected. Cryptocurrencies rely on blockchain technology to operate, as blockchain serves as the decentralized ledger that records all transactions and ensures their security and transparency. [7] As the internet becomes more accessible and convenient, an increasing number of people and organizations are turning to digital transactions. Digital payment systems are significantly faster, less expensive, and more efficient. As a result, it's not unexpected that innovative digital payment system types are quickly emerging. No other approach even comes close to the colossus that is cryptocurrencies. Predicting cryptocurrency prices can be useful for a variety of reasons. For traders and investors, predicting cryptocurrency prices can help them make informed decisions about when to buy or sell cryptocurrencies, maximizing their profits or minimizing their losses. For prediction, the algorithms used are GRU (gated recurrent unit), LSTM (longshort-term memory), and Bi-LSTM (Bi-directional long-short-term memory) algorithms to predict the future price of a cryptocurrency. An ensemble model is also created using the three models, and prices could be accurately predicted using these models and displaying the obtained results.

Keywords : Cryptocurrency; LSTM, Bi-LSTM, GRU, Ensambled;