Forecasting Bitcoin Prices Using Deep Learning for Consumer-Centric Industrial Applications
- 2 October 2023
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Consumer Electronics
- Vol. 70 (1) , 1351-1358
Abstract
As cryptocurrencies become more popular as investment vehicles, bitcoin draws interest from businesses, consumers, and computer scientists all across the world. Bitcoin is a computer file stored in digital wallet applications where each transaction is secured using strong cryptographic algorithms. It was challenging to forecast the future price of bitcoin due to its nonlinearity and extreme volatility. Several recent classic parametric models have been found with limited accuracy. To address the limitations and fill the existing research gaps, there is a need for a good prediction model which will provide the desired accuracy in the case of uncertainty and dynamism. This research suggested a deep learning-based framework for predicting and forecasting Bitcoin price. The research will be helpful for worldwide consumers and industries to take their decision on whether to invest or not. The research utilizes Yahoo! finance dataset for the period of 01-03-2016 to 26-02-2021 having 1828 samples. The experimental outcomes of the proposed Long Short-Term Memory (LSTM) model outperformed similar deep learning models by securing minimum loss and confirming that it can be used for future price prediction of the cryptocurrencies, which is helpful for the buyer to take their decision.Keywords
This publication has 31 references indexed in Scilit:
- Modal decomposition-based hybrid model for stock index predictionExpert Systems with Applications, 2022
- Stock price prediction using deep learning and frequency decompositionExpert Systems with Applications, 2021
- Deep Neural Network to Predict Answer Votes on Community Question Answering SitesNeural Processing Letters, 2021
- The volatility of Bitcoin and its role as a medium of exchange and a store of valueEmpirical Economics, 2021
- Bitcoin Price Prediction Using Time Series Analysis and Machine Learning Techniques2020
- Bitcoin Price Prediction and Analysis Using Deep Learning Models2020
- Bitcoin price prediction using machine learning: An approach to sample dimension engineeringJournal of Computational and Applied Mathematics, 2020
- Cryptocurrency forecasting with deep learning chaotic neural networksChaos, Solitons & Fractals, 2019
- The inefficiency of BitcoinEconomics Letters, 2016
- A novel hybridization of artificial neural networks and ARIMA models for time series forecastingApplied Soft Computing, 2011