Forecasting Market Closing Direction: A Novel Data Fusion of Deep News Embeddings and Technical Indicators with Gradient Boosting

Zare Baghiabad, Fatemeh; Karimi, Marzieh

doi:10.30699/ijf.2026.570396.1570

Forecasting Market Closing Direction: A Novel Data Fusion of Deep News Embeddings and Technical Indicators with Gradient Boosting

Document Type : Original Article

Authors

Fatemeh Zare Baghiabad

Marzieh Karimi

Assistant prof., Department of Industrial Engineering, Faculty of Engineering, Ardakan University, Ardakan, Iran.

10.30699/ijf.2026.570396.1570

Abstract

Accurately predicting stock market movements remains a complex engineering challenge due to inherent non-linearity and volatility. This study addresses this problem by developing a novel artificial intelligence framework that fuses deep learning-based natural language processing with technical analysis for financial forecasting. Its primary engineering application is predicting the daily directional movement of the Tehran Stock Exchange total index, a critical task for automated trading systems and risk management. The core AI contribution is a hybrid architecture that generates contextual text embeddings from Persian news using a Bidirectional Encoder Representations from Transformers (BERT) model and integrates them with technical indicators. These features are processed by a soft-voting ensemble of powerful gradient-boosting algorithms, namely eXtreme Gradient Boosting (XG-Boost) and Light Gradient Boosting Machine (LightGBM). Rigorously validated through an extensive walk-forward testing procedure across 331 temporal windows, the model achieved impressive performance metrics: 84.0% accuracy, 86.1% F1 Score, 87.0% precision, and 85.2% recall. These notable results demonstrate the significant value of domain-specific natural language processing in non-English financial contexts. Furthermore, this work establishes gradient boosting ensembles as a highly efficient, high-performance alternative to Long Short-Term Memory (LSTM) networks for financial forecasting. The proposed methodology demonstrates robust predictive capability for directional movement forecasting in emerging markets.

Keywords

Stock Market Forecasting

Persian Natural Language Processing

Bidirectional Encoder Representations from Transformers Embeddings

Gradient Boosting Ensemble

Multi-Modal Data Fusion

Chen, T., & Guestrin, C. (2016). Xgboost: A scalable tree boosting system. The 22nd Acm Sigkdd International Conference on Knowledge Discovery and Data Mining, 785–794. https://doi.org/10.1145/2939672.2939785

Devlin, J., Chang, M., Lee, K., & Toutanova, K. (2019). Bert: Pre-training of deep bidirectional transformers for language understanding. NAACL-HLT, 4171–4186. https://aclanthology.org/N19-1423/?utm_campaign=The+Batch&utm_source=hs_email&utm_medium=email&_hsenc=p2ANqtz-_m9bbH_7ECE1h3lZ3D61TYg52rKpifVNjL4fvJ85uqggrXsWDBTB7YooFLJeNXHWqhvOyC

Gurgul, V., Lessmann, S., & Härdle, W. K. (2025). Deep learning and NLP in cryptocurrency forecasting: Integrating financial, blockchain, and social media data. International Journal of Forecasting. https://doi.org/10.1016/j.ijforecast.2025.02.007

Hassan, M., Youssif, A., Imam, O., & Ghoneim, A. (2022). On the Impact of News for Reliable Stock Market Predictions: An LSTM-based Ensemble using FinBERT Word-Embeddings. WSEAS Transactions on Computers, 21, 294–303. https://wseas.com/journals/computers/2022/a725105-1432.pdf

Ke, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., Ye, Q., & Liu, T.-Y. (2017). Lightgbm: A highly efficient gradient boosting decision tree. Advances in Neural Information Processing Systems, 30. https://proceedings.neurips.cc/paper/2017/hash/6449f44a102fde848669bdd9eb6b76fa-Abstract.html

Li, M. (2024). Unraveling Financial Markets: Deep Neural Network-Based Models for Stock Price Prediction. Advances in Economics, Management and Political Sciences, 82(1), 195–203. https://doi.org/10.54254/2754-1169/82/20231111

Mao, Y., Yu, X., Zhu, J., & Wang, F. (2025). Forecasting crude oil futures price uncertainty: an explainable deep learning framework integrating news sentiment. Expert Systems with Applications, 297, 129419. https://doi.org/10.1016/j.eswa.2025.129419

Ohliati, J., & Yuniarty. (2024). Deep Learning for Stock Market Prediction: A Review. International Conference on Information Management and Technology, 666–671. https://doi.org/10.1109/ICIMTECH63123.2024.10780931

Oluwatobiloba, A. (2025). Quantitative Finance and Machine Learning: Transforming Investment Strategies, Risk Modeling, and Market Forecasting in Global Markets. International Journal of Computer Applications in Technology, 14(2319-8656), 34–49. https://www.researchgate.net/profile/Oluwatobiloba-Alao-2/publication/389498651_Quantitative_Finance_and_Machine_Learning_Transforming_Investment_Strategies_Risk_Modeling_and_Market_Forecasting_in_Global_Markets/links/67c4903d645ef274a499342e/Quantitative

Sarkar, S. K. (2023). Principal component analysis. Statistical Procedures for Analyzing Agricultural Data Using R, 8(5), 2028–2032. https://www.researchgate.net/profile/Santosha-Rathod/publication/391930391_Statistical_Procedures_for_Analyzing_Agricultural_Data_using_R/links/682da9098a76251f22e3f10b/Statistical-Procedures-for-Analyzing-Agricultural-Data-using-R.pdf#page=148

Takale, D. (2024). Enhancing Financial Sentiment Analysis: A Deep Dive into Natural Language Processing for Market Prediction Industries. Journal of Computer Networks and Virtualization, 2(03), 1–5. https://doi.org/10.48001/JOCNV.2024.221-5

Tan, Y. (2024). Analyzing trends in trading behaviour in financial markets using deep learning algorithms. International Journal of Information and Communication Technology, 25(10), 34–46. https://doi.org/10.1504/IJICT.2024.143332

Ti, Z. (2024). Stock Prediction using Deep Learning: A Comparison. Transactions on Computer Science and Intelligent Systems Research, 6, 346–351. https://doi.org/10.62051/B4FEFR32

Zhang, D., Tang, N., Dong, W., & Zhao, L. (2024). Machine Learning-Based Financial Big Data Analysis and Forecasting: From Preprocessing to Deep Learning Models. Applied and Computational Engineering, 100(1), 79–85. https://www.semanticscholar.org/paper/Machine-Learning-Based-Financial-Big-Data-Analysis-Zhang-Tang/244c63fd87a3d33dea7eac0a5bef97dfd2d434a4