AI in Cryptocurrency

【Author】 Iliev, Alexander I.; Panwar, Malvika

【Source】ADVANCES IN INFORMATION AND COMMUNICATION, FICC, VOL 2

【影响因子】

【Abstract】This study investigates the predictability of six significant cryptocurrencies for the upcoming two days using AI techniques i.e., machine learning algorithms like random forest and gradient for predicting the price of these six cryptocurrencies. The study presents to us that machine learning can be seen as a medium to predict the prices of cryptocurrencies. A machine learning system learns from past data, constructs prediction models, and predicts the output for new data whenever it gets it. Predicted output's accuracy is influenced by the quantity of data since the more data there is, the better the model can predict the output. The results show that with the accuracy score performance metric, which we employed for this study, we were able to calculate the accuracy of the algorithms and find that both algorithms random forest and gradient boosting respectively performed well for the cryptocurrencies such as Solana (98.07%,98.14%), Binance (96.56%, 96.85%), and Ethereum (96.61%, 96.60%)), with the exception of Tether (0.38%, 12.35%) and USD coin (-0.59%, 1.48%), the results demonstrate that both algorithms work effectively with themajority of cryptocurrencies which can be further increased by using deep learning algorithms like ANN, RNN or LSTM.

【Keywords】Cryptocurrency; Artificial intelligence; Machine learning; Deep learning; ANN; RNN; LSTM

【发表时间】2023

【收录时间】2023-07-17

【文献类型】实证数据

【主题类别】

区块链治理-市场治理-价格预测