High-Frequency Stock Market Price Prediction Using Blockchain and Deep Learning

Researchers and financiers utilize statistics on stock market operation, including past returns, share values, accessible, large, poor, near, and transaction size, to identify trends in stock activity. Technological evaluation can provide financiers with contradictory indications. They focused on predicting the return and risk levels of stock market prices. As a result, the stock market requires an automatic technology that can anticipate the return rate of bitcoin. Deep learning (DL) technology can estimate share market trends employing technological evaluation. This research creates a deep learning-based smart return rate estimation method for blockchain-based stock market prices (RRE-DLBSP). The proposed RRE-DLBSP framework develops an LSTM (long short-term memory) paradigm for RRE assessment. Additionally, the LSTM algorithm's hyperparameters are effectively adjusted using the Adam optimization algorithm, improving prediction efficiency. The Ethereum return rate is selected as the goal and the model outcomes are examined in various ways to ensure the highest functioning of the RRE-DLBSP paradigm. The research results demonstrated the RRE-DLBSP method's superiority over the cutting-edge approaches based on a variety of performance metrics. The suggested RRE-DLBSP has an MSE of 0.0436 and 0.0656, while the mean for the evaluated techniques in training and evaluation is 0.725 and 0.6124, respectively.