IRO Journals
Journal of Soft Computing Paradigm

Bitcoin Price Prediction Using Machine Learning Technique

Open Access
Published: 17 October, 2026
Pages: 388-403
  • Sujith V. , Sujith V.
    Assistant Professor, Department of Computer Science and Engineering, NPR College of Engineering and Technology, Dindigul, India
    Assistant Professor, Department of Computer Science and Engineering, NPR College of Engineering and Technology, Dindigul, India
  • Sree Yogeswari S. , Sree Yogeswari S.
    Student, Department of Computer Science and Engineering, NPR College of Engineering and Technology, Dindigul, India
    Student, Department of Computer Science and Engineering, NPR College of Engineering and Technology, Dindigul, India
  • Subha Sri R. Subha Sri R.
    Student, Department of Computer Science and Engineering, NPR College of Engineering and Technology, Dindigul, India
    Student, Department of Computer Science and Engineering, NPR College of Engineering and Technology, Dindigul, India
view PDF
view PDF

How to Cite

V., Sujith, Sree Yogeswari S., and Subha Sri R. 2026. “Bitcoin Price Prediction Using Machine Learning Technique”. Journal of Soft Computing Paradigm 8 (4): 388-403. https://doi.org/10.36548/jscp.2026.4.005.
ACM ACS APA ABNT Chicago Harvard IEEE MLA Turabian Vancouver AMA

Download Citation

Endnote/Zotero/Mendeley (RIS) BibTeX

Keywords

Bitcoin Price Forecasting
XGBoost Regression
Support Vector Regression
Random Forest Ensemble
Prediction Fusion
Market Trend Classification
Flask-Based Predictive Analytics

Abstract

Bitcoin price forecasting is challenging because of nonlinear market behaviour and rapid price fluctuations. This study presents an integrated machine-learning framework for one-step-ahead Bitcoin price prediction using Random Forest (RF), Support Vector Machine (SVM), and XGBoost. The framework incorporates preprocessing, comparative model evaluation, prediction fusion, and directional trend classification within a unified predictive workflow. Model performance is assessed using MAE, MSE, RMSE, and R², while an equal-weight fusion mechanism combines the individual prediction outputs. A separate RF classifier generates UP, DOWN, and NEUTRAL market-direction states. Experimental results demonstrate that XGBoost provides the strongest predictive performance, achieving a reported accuracy of 96.5% and an R² of 0.94, with lower prediction errors than the other evaluated models. The complete framework is deployed through a Flask-based web application supporting interactive prediction, graphical visualization, trend analysis, and prediction-history management.

References

  1. Velankar, Siddhi, Sakshi Valecha, and Shreya Maji. "Bitcoin price prediction using machine learning." In 20th International Conference on Advanced Communication Technology (ICACT), IEEE, 2018, 144-147.
  2. Srivastava, Vibha, Vijay Kumar Dwivedi, and Ashutosh Kumar Singh. "Cryptocurrency price prediction using enhanced PSO with extreme gradient boosting algorithm." Cybernetics and Information Technologies, 2023, vol. 23, no. 2: 170-187.
  3. Aggarwal, Divya, Shabana Chandrasekaran, and Balamurugan Annamalai. "A complete empirical ensemble mode decomposition and support vector machine-based approach to predict Bitcoin prices." Journal of Behavioral and Experimental Finance, 2020, vol. 27: 100335.
  4. Ganesan, Shanmuga Canaraj, Saripelli Chaitanya Bangarraju, Kishan Tiwari, Katakam Jaswanthi, and Amiya Bhaumik. "Bitcoin Price Prediction Using XGBoost Classifier: A Machine Learning Approach." Soft Computing and Signal Processing: Proceedings of 7th ICSCSP, 2024, vol. 1, no. 1: 91.
  5. Nielsen, Michael A. "Neural networks and deep learning." (2015).
  6. Wang, Sidan, Zidong Huang, and Zhong-Wei Liao. "Multi-scale decomposition for deep learning-based Bitcoin price forecasting." Finance Research Letters, 2026, vol. 96: 109776.
  7. Chen, Tianqi, and Carlos Guestrin. "Xgboost: A scalable tree boosting system." In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2016, 785-794.
  8. Box, George EP, Gwilym M. Jenkins, Gregory C. Reinsel, and Greta M. Ljung. Time series analysis: forecasting and control. John Wiley & Sons, 2015.
  9. Brownlee, Jason. Machine learning mastery. Machine Learning Mastery, 2022.
  10. Géron, Aurélien. Hands-on machine learning with Scikit-Learn, Keras, and TensorFlow. " O'Reilly Media, Inc.", 2022.
  11. Yahoo Finance API, Dataset, Accessed 2025. Available: https://finance.yahoo.com/