Abstract

Time-series forecasting under non-stationary and complex conditions is still an unresolved problem in machine learning, especially when the information of interest is spread across non-homogeneous sources of data. Conventional statistical methods, such as Autoregressive Integrated Moving Average (ARIMA) models, rely on restrictive assumptions and are often unable to capture nonlinear temporal dynamics, while recurrent neural networks, such as Long Short-Term Memory (LSTM) models, may fail to adequately represent distributional properties under volatile conditions. This study proposes a hybrid architectural framework that integrates Generative Adversarial Networks (GANs) with Natural Language Processing (NLP) techniques for sentiment extraction to improve time-series prediction. The proposed architecture combines adversarial learning on numerical market data with sentiment-based contextual information derived from unstructured textual sources, allowing the model to jointly exploit temporal and contextual signals. The framework is evaluated on financial market data, a representative application domain characterized by noise, regime shifts, and heterogeneous information streams. Experimental results demonstrate that the proposed GAN–NLP approach outperforms ARIMA and LSTM baselines across multiple evaluation metrics and exhibits improved robustness under varying volatility regimes. The findings highlight the potential of sentiment-conditioned generative modelling as a flexible and effective approach for time-series prediction tasks involving both numerical and textual data.

References

• Billings, D. and Yang, J.-S. 2006. Application of the ARIMA models to urban roadway travel time prediction: a case study. In Proceedings of IEEE International Conference on Systems, Man and Cybernetics, pp. 2529–2534.
• Engle, R. F. 1982. Autoregressive conditional heteroscedasticity with estimates of the variance of United Kingdom inflation. Econometrica, 50(4), pp. 987–1007.
• Hochreiter, S. and Schmidhuber, J. 1997. Long short-term memory. Neural Computation, 9(8), pp. 1735–1780.
• Bengio, Y., Simard, P., and Frasconi, P. 1994. Learning long-term dependencies with gradient descent is difficult. IEEE Transactions on Neural Networks, 5(2), pp. 157–166.
• Fischer, T. and Krauss, C. 2018. Deep learning with long short-term memory networks for financial market predictions. European Journal of Operational Research, 270(2), pp. 654–669.
• Nelson, D. M. Q., Pereira, A. C. M., and de Oliveira, R. A. 2017. Stock market’s price movement prediction with LSTM neural networks. In Proceedings of International Joint Conference on Neural Networks, pp. 1419–1426.
• Goodfellow, I. J., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., and Bengio, Y. 2014. Generative adversarial networks. arXiv preprint.
• Zhang, K., Zhong, G., Dong, J., Wang, S., and Wang, Y. 2019. Stock market prediction based on generative adversarial network. Procedia Computer Science, 147, pp. 400–406.
• Vuletić, M., Prenzel, F., and Cucuringu, M. 2023. Fin-GAN: forecasting and classifying financial time series via generative adversarial networks. SSRN Electronic Journal.
• Brophy, E., Wang, Z., She, Q., and Ward, T. E. 2023. Generative adversarial networks in time series: a systematic literature review. ACM Computing Surveys, 55(10).
• Bollen, J., Mao, H., and Zeng, X. 2011. Twitter mood predicts the stock market. Journal of Computational Science, 2(1), pp. 1–8.
• Derakhshan, A. and Beigy, H. 2019. Sentiment analysis on stock social media for stock price movement prediction. Engineering Applications of Artificial Intelligence, 85, pp. 569–578.
• Puh, K. and Bagić Babac, M. 2023. Predicting stock market using natural language processing. American Journal of Business, 38(2), pp. 41–61.
• Tuan, N. M., Meesad, P., and Son, N. H. 2024. On a stock prediction aligned to natural language sentiments. In Proceedings of International Conference on Natural Language Processing and Information Retrieval, Article 111.
• Dedotsi, S., Lazanas, A., Siachos, I., Teloni, D. D., and Telonis, A. G. 2023. Discrete clusters formulation through the exploitation of optimized k-modes algorithm for hypotheses validation in social work research: the case of Greek social workers working with refugees. BOHR International Journal of Internet of Things, Artificial Intelligence and Machine Learning, 2(1), pp. 11–18.
• Box, G. E. P., Jenkins, G. M., Reinsel, G. C., and Ljung, G. M. 2015. Time Series Analysis: Forecasting and Control. 5th ed. Wiley, Hoboken, NJ.
• Fama, E. F. 1970. Efficient capital markets: a review of theory and empirical work. The Journal of Finance, 25(2), pp. 383–417.
• Kahneman, D. and Tversky, A. 1979. Prospect theory: an analysis of decision under risk. Econometrica, 47(2), pp. 263–291.
• Shiller, R. J. 2015. Irrational Exuberance. 3rd ed. Princeton University Press, Princeton, NJ.
• Turney, P. D. 2002. Thumbs up or thumbs down? Semantic orientation applied to unsupervised classification of reviews. In Proceedings of 40th Annual Meeting of the Association for Computational Linguistics, pp. 417–424.
• Whissell, C. M. 1989. The dictionary of affect in language. In Plutchik, R. and Kellerman, H. (eds.), The Measurement of Emotions, Elsevier, pp. 113–131.
• Turney, P. D. and Littman, M. L. 2003. Measuring praise and criticism. ACM Transactions on Information Systems, 21(4), pp. 315–346.
• Leung, C. K. S., MacKinnon, R. K., and Wang, Y. 2014. A machine learning approach for stock price prediction. In Proceedings of 18th International Database Engineering and Applications Symposium, pp. 274–277.
• Shen, J. and Shafiq, M. O. 2020. Short-term stock market price trend prediction using a comprehensive deep learning system. Journal of Big Data, 7(1).
• Sonkiya, P., Bajpai, V., and Bansal, A. 2021. Stock price prediction using BERT and GAN. arXiv preprint.
• Xu, Z., Bian, Y., Zhong, J., Wen, X., and Xu, Q. 2024. Beyond trend and periodicity: guiding time series forecasting with textual cues. arXiv preprint.
• Emami, H., Dang, X.-H., Shah, Y., and Zerfos, P. 2023. Modality-aware transformer for financial time series forecasting. arXiv preprint.
• Moreno, A. and Ordieres-Meré, J. 2025. Predicting stock price trends using language models to extract the sentiment from analyst reports: evidence from IBEX 35-listed companies. Economics Letters.
• Kim, C., Kim, S., Kim, J., Lee, D., and Kim, S. 2021. Automated learning rate scheduler for large-batch training. arXiv preprint.
• Prechelt, L. 2012. Early stopping — but when? In Montavon, G., Orr, G. B., and Müller, K.-R. (eds.), Neural Networks: Tricks of the Trade. 2nd ed. Springer, pp. 53–67.
• Zhang, C., Sjarif, N. N. A., and Ibrahim, R. 2024. Deep learning models for price forecasting of financial time series: a review of recent advancements (2020–2022). WIREs Data Mining and Knowledge Discovery, 14(1).
• Rouf, N. et al. 2021. Stock market prediction using machine learning techniques: a decade survey on methodologies, recent developments, and future directions. Applied Sciences, 10(21).
• Benidis, K. et al. 2022. Deep learning for time series forecasting: tutorial and literature survey. ACM Computing Surveys.
• Julian, T., Devrison, T., Anora, V., and Suryaningrum, K. M. 2023. Stock price prediction model using deep learning optimization based on technical analysis indicators. Procedia Computer Science, 227, pp. 939–947.
• Li, S. 2025. Enhancing stock price prediction using GANs and transformer-based attention mechanisms. Empirical Economics.
• Woo, J. et al. 2025. Towards time series generation conditioned on unstructured natural language descriptions. arXiv preprint.
• Azar, P. 2016. The wisdom of Twitter crowds: predicting stock market reactions to FOMC meetings via Twitter feeds. SSRN Electronic Journal.
• Goodfellow, I. 2016. NIPS 2016 tutorial: generative adversarial networks. arXiv preprint