Skip to content
Open access

A hybrid FinBERT-LSTM framework for Bitcoin price forecasting using news sentiment and technical indicators

Jul 2026 · Balıkesir Üniversitesi Fen Bilimleri Enstitüsü Dergisi · 0 citations · 9 references

TL;DR

A hybrid forecasting framework that integrates sentiment analysis with deep learning to predict Bitcoin’s hourly and daily closing prices and empirical results demonstrate that sentiment-enhanced hybrid models consistently outperform models based solely on technical indicators across RMSE, MAE, MAPE, and R² metrics.

Abstract

This study proposes a hybrid forecasting framework that integrates sentiment analysis with deep learning to predict Bitcoin’s hourly and daily closing prices. Hourly BTC/USD market data spanning June 2021 to November 2025 were combined with approximately 326,000 Bitcoin-related news headlines published over the same period. Sentiment scores in the range of [-1, +1] were generated for each headline using FinBERT, a transformer-based language model trained on financial texts, and were subsequently integrated with technical indicators such as trading volume, MACD, and RSI. The resulting combined feature set was modeled using an LSTM network to capture temporal dependencies. Empirical results demonstrate that sentiment-enhanced hybrid models consistently outperform models based solely on technical indicators across RMSE, MAE, MAPE, and R² metrics. The hourly hybrid model achieved the best performance, with an RMSE of 1,009 USD and an R² of 99.23%. Furthermore, a 30-day out-of-sample real-time evaluation yielded an RMSE of 941 USD. The consistency between in-sample and out-of-sample results indicates that the proposed framework maintains stable predictive performance over time.

Read PDF

Similar papers

Open access Aug 2026

FORECASTING BITCOIN PRICE RANGES BY COMBINING SENTIMENT ANALYSIS WITH HIGH-DIMENSIONAL INDICATORS IN AN LSTM FRAMEWORK

A way to estimate the range of Bitcoin prices for the following day by combining Long Short-Term Memory (LSTM) networks with natural language processing (NLP) approaches, showing how sentiment analysis and deep learning architectures can be combined to increase forecasting resilience and interpretability in erratic cryptocurrency markets.

Hajera Fatima, C. B. Jones · 0 citations
Review Open access Jul 2026

Indian Stock Market Forecasting Using LSTM-XGBoost and Technical Indicators

The Indian stock market is characterized by high volatility, non-linear price behaviour, and sensitivity to macroeconomic, sectoral, and sentiment-driven factors, which limits the accuracy of traditional linear forecasting models such as ARIMA. Building on our earlier literature review and problem formulation, this paper presents the implementation and evaluation of an integrated deep learning framework for next-day closing price prediction of Indian equities. The framework combines a two-layer Long Short-Term Memory (LSTM) network with four complementary technical indicators — Moving Average Convergence Divergence (MACD), Relative Strength Index (RSI), the 10–20 day Exponential Moving Average (EMA) crossover, and the Stochastic Oscillator — as engineered input features, and a downstream XGBoost classifier that converts the LSTM's continuous price forecast, together with the current indicator states, into discrete BUY, HOLD, or SELL trading signals with associated confidence scores. The complete pipeline is implemented as a full-stack platform (Python, Flask, MongoDB, React) that retrieves real-time NSE/BSE data through the yfinance API. The proposed model is evaluated on RELIANCE.NS using Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE), supported by a per-indicator ablation study and a comparison against an autoregressive (AR) linear baseline, a Support Vector Machine (SVM) regressor, and a single-indicator LSTM baseline. The proposed model achieved an RMSE of ₹28.56 and MAE of ₹23.48 (MAPE 1.71%) on the held-out test partition, achieving the lowest RMSE among all compared models, while the downstream XGBoost signal classifier achieved 91.7% accuracy on held-out BUY/HOLD/SELL labels.

Ayush Jha, Pankaj Singh · 0 citations
Open access Aug 2026

Applying News Sentiment Analysis to Bitcoin Price Prediction

Bitcoin price forecasting remains difficult because cryptocurrency markets exhibit volatility, nonlinear temporal dependence, and rapid responses to external information. This study examines whether news-derived sentiment improves a Gated Recurrent Unit (GRU) model beyond historical market variables alone. Bitcoin open, high, low, close, and volume data from Binance and Bitcoin-related news from btcnews.com covered January 2020–December 2025. After relevance filtering, 12,332 articles were processed using entity-focused Aspect-Based Sentiment Analysis and FinVADER to generate mean polarity, polarity dispersion, and article-volume features. Two identical two-layer GRU models used the preceding 30-day sequence to predict the next day’s closing price: a baseline with five market variables and an enhanced model with three sentiment features. News was aggregated by publication date without an explicit market-close cut-off; therefore, look-ahead effects cannot be excluded. Data were split chronologically into 70% training, 15% validation, and 15% testing, with scaling fitted only on the training set. In the reported single run, the enhanced model achieved RMSE USD 2,052.65, MAE USD 1,525.67, MAPE 1.51%, and R² 0.9728, compared with USD 2,244.68, USD 1,743.90, 1.70%, and 0.9675 for the baseline. Across five runs, mean RMSE improved from USD 2,402.34 to USD 2,238.46, although variability was greater for the enhanced model. A one-day-ahead Diebold–Mariano test on the selected-run test-error series indicated a significant difference (DM = 3.3605; p = 0.0008), conditional on the reported error construction. Overall, news sentiment provided complementary predictive information, while temporal alignment and initialization sensitivity require further validation.

Ahmad Hauzan Abid Rommadhon, Rizky Parlika, Achmad Junaidi · 0 citations
Open access Sep 2026

Integrating FinBERT based news sentiment and technical indicators for LSTM stock price forecasting

Accurate stock price forecasting is important for investment and risk management but remains challenging due to complex interactions among market dynamics, firm-specific information, and investor sentiment. Although technical indicators capture historical price patterns, they may not fully reflect information conveyed through financial news. This study develops an LSTM-based framework that integrates technical indicators with FinBERT-derived news sentiment for next-day stock price forecasting. The model uses closing price, EMA26, MACD, RSI14, and daily sentiment scores extracted from 58,466 GDELT news headlines for Amazon, Apple, Google, Microsoft, and NVIDIA from 2019 to 2023. Company-level sentiment is temporally aligned with market features, and 15-day sequences are constructed to predict the normalized next-trading-day closing price. Out-of-sample evaluation from July to December 2023 reveals heterogeneous performance across firms: Apple achieves the lowest RMSE and MAE (0.0147 and 0.0110), while NVIDIA records the highest (0.0245 and 0.0201). Predictions generally capture temporal price dynamics but are less responsive to abrupt turning points. The findings demonstrate a systematic approach to integrating financial news sentiment and technical indicators for stock price forecasting while highlighting substantial cross-firm variation in predictive performance.

Yu-Zheng Zhao · 0 citations
Open access Jul 2026

Crypto Price Analysis: An AI Perspective

A methodological framework based on LSTM neural networks for comparing the performance of various sets of features in predicting Bitcoin prices and proof that technical indicators have a significant impact on the predictive accuracy of this model, while social media sentiment has a minor impact under the circumstances are brought.

Sedeeq Hasan Banna, Ammar Ahmed Othman, Khaled Al-Raddah · 0 citations
Open access Aug 2026

Comparative Analysis of LSTM and GRU Models with Hyperparameter Optimization for Bitcoin Price Prediction

This work attempts to provide a thorough comparative analysis mapping the precise accuracy–efficiency trade-off between Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) models under a standardized Grid Search hyperparameter optimization pipeline using a recent Bitcoin closing-price dataset.

Moh Quthbul Widad, Rizky Parlika, Firza Prima Aditiawan · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.