Adaptive Hybrid Learning for Real-Time Bitcoin Price Prediction with Comparative Model Selection
Journal of Artificial Intelligence and Capsule Networks · 2026
DOI: 10.36548/aicn.2026.4.006 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Artificial Intelligence and Capsule Networks, 2026. Verfasst von Sai Vedikaa R. und 5 weiteren Autor:innen. Die Arbeit nennt 21 Literaturangaben.
Abstract (Originalsprache)
This study presents a real-time Bitcoin price forecasting framework that integrates Long Short-Term Memory (LSTM), Improved Sequential Temporal Modeling (ISTM), gradient boosting, and Flask-based deployment. The experimental dataset comprises 744 hourly BTC/USDT Spot Kline observations obtained from the Binance Public Data Repository, including Open, High, Low, Close, and Volume (OHLCV) variables. A set of technical indicators and temporal features, including moving averages, exponential moving averages, Relative Strength Index (RSI), volatility measures, and lagged variables, was derived from the historical price series. Following Min-Max normalization, the dataset was chronologically divided into 595 training observations and 149 testing observations using an 80:20 split without shuffling to preserve the temporal structure of the data. Six forecasting approaches such as Linear Regression, Random Forest, LSTM, ISTM, LSTM with Boosting, and ISTM with Boosting. Among the evaluated models, the ISTM + Boosting framework demonstrated strong predictive performance, achieving an MAE of 72.5, MSE of 19,860, RMSE of 140.9, and R² of 0.97.
Zitieren
APA
R., S. V., V., S. S., A., K. R., K., S., P., Y., & S., S. R. L. N. V. (2026). Adaptive Hybrid Learning for Real-Time Bitcoin Price Prediction with Comparative Model Selection. Journal of Artificial Intelligence and Capsule Networks. https://doi.org/10.36548/aicn.2026.4.006
Vancouver
R. SV, V. SS, A. KR, K. S, P. Y, S. SRLNV. Adaptive Hybrid Learning for Real-Time Bitcoin Price Prediction with Comparative Model Selection. Journal of Artificial Intelligence and Capsule Networks. 2026. doi:10.36548/aicn.2026.4.006
BibTeX
@article{R2026,
author = {R., Sai Vedikaa and V., Siva Srinidhi and A., Keerthika Rupa and K., Srilekha and P., Yuvashree and S., Sree Rathna Lakshmi N. V.},
title = {Adaptive Hybrid Learning for Real-Time Bitcoin Price Prediction with Comparative Model Selection},
journal = {Journal of Artificial Intelligence and Capsule Networks},
year = {2026},
doi = {10.36548/aicn.2026.4.006}
}
EUVIMED