Real-time Cryptocurrency Market Prediction using News Sentiment Analysis and Machine Learning
Abstract
The increasing size of cryptocurrency markets creates a requirement for advanced systems that can perform market sentiment analysis and create price forecasts for short periods. The study introduces a crypto market intelligence system that operates in real-time and combines news sentiment analysis with high-frequency price data to predict market movements over brief time periods. The system obtains news information through the CryptoPanic API and acquires minute-by-minute cryptocurrency price information from Binance. The VADER natural language processing model enables computation of sentiment polarity scores, which help assess positive, negative, and neutral news events. The proposed system aligns news timestamps with corresponding market prices and extracts engineered features, including volatility, momentum, and percentage price change within a one-hour window. The system uses Random Forest regression to develop a machine learning model that predicts short-term price changes as news sentiment affects market prices. The experimental results showed that combining textual sentiment indicators with quantitative trading signals enables accurate short-term cryptocurrency price predictions. The developed framework enables automated data acquisition and feature extraction, model development, and real-time crypto analysis through its dashboard interface, which provides various interactive tools. The best prediction accuracy is achieved by hybrid deep learning models. The algorithm can predict price movement with an accuracy of 79%, which confirms the possibility of using news headlines to improve short-term predictions in cryptocurrency markets. The AI models may be trained using measurements of sentiment retrieved from financial news and technical data from the market to predict price movements. Particularly, the influence of news can be predicted by identifying and analyzing it within an hour of publication. Advanced systems are currently going from daily forecasts to hourly, short-term forecasting. High-frequency tracking makes it beneficial for addressing issues like price slippage. Experimental data are provided to demonstrate the usefulness of the developed categorization model.
References
S. M. Raju and A. M. Tarif, “Real-time prediction of BITCOIN price using machine learning techniques and public sentiment analysis,” arXiv, Jun. 2020.
S. Mohapatra, N. Ahmed, and P. Alencar, “KryptoOracle: A real-time cryptocurrency price prediction platform using Twitter sentiments,” 2019 IEEE International Conference on Big Data (Big Data), Los Angeles, CA, USA, 2019, pp. 5544–5551.
F. Feizian and B. Amiri, “Cryptocurrency price prediction model based on sentiment analysis and social influence,” IEEE Access, vol. 11, pp. 142177–142195, Jan. 2023.
D. Perera, J. Lim, S. Gunraku, and W. Lim, “Cryptocurrency analysis: Price prediction of cryptocurrency using user sentiments and quantitative data,” Proceedings of the 16th International Conference on Agents and Artificial Intelligence - Volume 3, Jan. 2024, pp. 210–217.
S. Singh and M. Bhat, “Transformer-based approach for Ethereum price prediction using crosscurrency correlation and sentiment analysis,” arXiv, Jan. 2024.
S. M. Dashtaki, R. M. Dashtaki, M. H. Chagahi, B. Moshiri, and M. J. Piran, “A multisource fusion framework for cryptocurrency price movement prediction,” arXiv.org, 2024.
S. Alghamdi, S. Alqethami, T. Alsubait, and H. Alhakami, “Cryptocurrency price prediction using forecasting and sentiment analysis,” International Journal of Advanced Computer Science and Applications, vol. 13, no. 10, 2022,
A. Y. Nouira, M. Bouchakwa, and Y. Jamoussi, “Bitcoin price prediction considering sentiment analysis on Twitter and Google News,” Proceedings of the 27th International Database Engineered Applications Symposium, May 2023, pp. 71–78.
S. Arslan, “Bitcoin price prediction using sentiment analysis and empirical mode decomposition,” Computational Economics, vol. 65, pp. 2227–2248, May 2024,
S. Simha, R. Devireddy, and H. Niculescu, Cryptocurrency Price Prediction Using Ensemble Methods and Sentiment Analysis, M.Sc. research project, Master's in Data Analytics, National College of Ireland, Dublin, Ireland.
S. Oikonomopoulos, K. Tzafilkou, D. Karapiperis and V. Verykios, “Cryptocurrency price prediction using social media sentiment analysis,” 2022 13th International Conference on Information, Intelligence, Systems & Applications (IISA), Corfu, Greece, 2022, pp. 1-8.
S. E. Parameswaran, V. Ramachandran, and S. Shukla, “Crypto trend prediction based on wavelet transform and deep learning algorithm,” Procedia Computer Science, vol. 235, pp. 1179–1189, May 2024.
R. Bangroo, U. Gupta, R. Sah and A. Kumar, “Cryptocurrency price prediction using machine learning algorithm,” 2022 10th International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions) (ICRITO), Noida, India, 2022, pp. 1-4.
H. Kavitha, U. K. Sinha and S. S. Jain, “Performance evaluation of machine learning algorithms for bitcoin price prediction,” 2020 Fourth International Conference on Inventive Systems and Control (ICISC), Coimbatore, India, 2020, pp. 110-114.
C. Li, “Predicting bitcoin prices using sentiment analysis results,” Proceedings of the 6th International Conference on E-Commerce, E-Business and E-Government, Jul. 2022, pp. 217–222.
