Analyzing Cryptocurrencys Markets Using Artificial Neural Network

Abstract

This article presents an analysis of cryptocurrency markets, which are popular for investors nowadays. The primary objectives of the present study were to examine correlation between individual cryptocurrency value and investors and to analyze cryptocurrency markets using machine learning algorithms with neural networks for correlational analysis of future cryptocurrency value trend. The results of digital currency market analysis with a neural network showed that the correlation coefficient for each cryptocurrency ranged between -1 and 1. The negative correlation indicates that two currencies did not have relationship with value fluctuation, while the positive value suggests that two currencies were correlated with value fluctuation. The mean squared error (MSE) of 0.008 was achieved to ensure the accuracy of crypto currencies market trends. It is highly recommended that include information of world market trends be considered in further analyses.