Advanced Use of Blockchain And Deep Learning Technologies For Financial Forecasting

Abstract

Blockchain technology dependent on cryptocurrency have lately excited the curiosity of capitalists. They focused on forecasting the financial item's risk and return ratios. As a result, financial items require an autonomous algorithm to anticipate the return percentage of cryptocurrency. Deep learning (DL) algorithms that were lately developed lay the path for the return percentage forecasting procedure. This paper proposes a blockchain financial product utilizing DL for a smart return rate prediction (RRP-DLBFP) method. The suggested RRP-DLBFP method entails creating a long short-term memory (LSTM) framework for return percentage forecasting. Furthermore, the Adam optimization is used to effectively change the LSTM algorithm's hyperparameters, resulting in improved forecasting accuracy. The Ethereum rate of return has been selected as the aim of guaranteeing the RRP-DLBFP method's superior performance, and its outcomes are studied in various metrics. In terms of several assessment variables, the model's results demonstrated the superiority of the RRP-DLBFP method over the present latest methods. The suggested RRP-DLBFP exhibits MSE values of 0.0435 & 0.0655, accordingly, contrasted with a mean of 0.6139 & 0.723 for comparing techniques in both training and evaluation.