Year: 2023
Type: article
Abstract: Cryptocurrencies, such as Bitcoin and Ethereum, have experienced significant price volatility over the years, and investors and traders often look for ways to predict future price movements to make informed investment decisions. However, predicting the prices of cryptocurrencies is a challenging task due to the highly unpredictable nature of the market and the lack of a centralized authority to regulate it. Overall, smart consensus algorithms play a crucial role in maintaining the security and reliability of decentralized systems by enabling all nodes to agree on the state of the network without the need for a centralized authority. Because of the problem of making predictions on the prices of cryptocurrencies, this system proposed a Bi-Directional Long Short-Memory algorithm for the prediction of bitcoin prices. This system uses stock market data starting from 2014 to 2022. The dataset was pre-processed so that it will be suitable for training a robust model. The model was trained using Bi-LSTM. The result of the model is promising with a Mean Absolute error of 0.012% and a predicting accuracy of 99.9%. The proposed system was compared with other existing models, and the result shows that the model outperforms the existing model. The proposed system model was also saved and deployed to the web so that users can make use of it in making a future prediction of the prices of cryptocurrencies.
Authors: J.B. Bekele, O.E. Taylor
Language: English
Field: Decision Sciences
Domain: Social Sciences
Open Access status: hybrid
OpenAlex ID: W4407445554