Market Trend Discovery Through Deep ‎Learning-Driven Twitter Sentiment for ‎Stock Forecasting

  • Authors

    • Mr. Lukesh Kadu Research Scholar, Computer Engineering Department, A. C. Patil College of Engineering, Navi Mumbai
    • Dr. M. M. Deshpande HOD, Computer Engineering Department, A. C. Patil College of Engineering, Kharghar, Navi Mumbai
    • Dr. V. N. Pawar Principal, A. C. Patil College of Engineering, Kharghar, Navi Mumbai
    https://doi.org/10.14419/c1qpgz81

    Received date: October 11, 2025

    Accepted date: November 2, 2025

    Published date: November 15, 2025

  • Sentiment Analysis; Deep Learning; Stock Pre-Diction; Twitter; LSTM; BERT

  • Abstract

    Short-range stock price forecasting presents considerable challenges due to the multifaceted and rapidly evolving nature of market signals. ‎Twitter, as a social media platform, delivers immediate public sentiment, serving as a valuable complement to traditional price-based metrics. This research introduces an operational framework merging Twitter sentiment classification with time-series forecasting that integrates ‎both sentiment and technical market features. The sentiment analysis component employs an ensemble approach combining LSTM, CNN, ‎and fine-tuned BERT architectures. These sentiment indicators feed into a neural network designed for time-series analysis to forecast directional price movements. Testing on datasets from January 2020 through December 2022 yielded 87.3% accuracy in sentiment classification ‎and 82.1% accuracy in directional prediction. These outcomes indicate that well-processed social media sentiment enhances short-term trading signals when combined with conventional technical indicators.

  • References

    1. Bollen, H. Mao, and X. Zeng, “Linking public mood on Twitter with movements in the stock market,” J. Comput. Sci., vol. 2, no. 1, pp. 1–8, 2011. https://doi.org/10.1016/j.jocs.2010.12.007.
    2. X. Zhang, H. Fuehres, and P. A. Gloor, “Forecasting market signals using Twitter discussions,” Procedia–Social Behav. Sci., vol. 26, pp. 55–62, 2018. https://doi.org/10.1016/j.sbspro.2011.10.562.
    3. A. Kumar and V. Ravi, “Applications of text mining in finance: A survey,” Knowl.-Based Syst., vol. 114, pp. 128–147, 2016. https://doi.org/10.1016/j.knosys.2016.10.003.
    4. D. Shah, H. Isah, and F. Zulkernine, “A taxonomy and review of stock prediction methodologies,” Int. J. Financ. Stud., vol. 7, no. 2, p. 26, 2019. https://doi.org/10.3390/ijfs7020026.
    5. J. Devlin, M.-W. Chang, K. Lee, and K. Toutanova, “BERT: Deep bidirectional transformers for language understanding,” in Proc. NAACL- HLT, 2019, pp. 4171–4186. https://doi.org/10.18653/v1/N19-1423.
    6. Y. Li, S. Pan, K. Zhang, Y. Yin, and Q. Chen, “Financial sentiment learn- ing for stock return prediction,” IEEE Access, vol. 8, pp. 78899–78908, 2020.
    7. T. Chen, R. Xu, Y. He, and X. Wang, “Enhancing sentiment analysis via sentence-type features with BiLSTM-CRF and CNN,” Expert Syst. Appl., vol. 72, pp. 221–230, 2021. https://doi.org/10.1016/j.eswa.2016.10.065.
    8. S. Hochreiter and J. Schmidhuber, “Long short-term memory networks, ”Neural Comput., vol. 9, no. 8, pp. 1735–1780, 1997. https://doi.org/10.1162/neco.1997.9.8.1735.
    9. Y. Kim, “CNNs for sentence-level text classification,” in Proc. EMNLP, 2014, pp. 1746–1751. https://doi.org/10.3115/v1/D14-1181.
    10. A. Vaswani et al., “Attention mechanisms in neural networks: The Transformer model,” in Proc. NeurIPS, 2017, pp. 5998–6008.
    11. R. Socher et al., “Recursive deep learning for sentiment treebank analysis,” in Proc. EMNLP, 2013, pp. 1631–1642. https://doi.org/10.18653/v1/D13-1170.
    12. M. E. Peters et al., “ELMo: Deep contextualized word representations,” in Proc. NAACL-HLT, 2018, pp. 2227–2237. https://doi.org/10.18653/v1/N18-1202.
    13. Z. Yang et al., “Hierarchical attention networks for document-level text classification,” in Proc. NAACL-HLT, 2016, pp. 1480–1489.
    14. X. Li, H. Xie, L. Chen, J. Wang, and X. Deng, “Impact of news sentiment on stock returns,” Knowl.-Based Syst., vol. 69, pp. 14–23, 2014. https://doi.org/10.1016/j.knosys.2014.04.022.
    15. P. C. Tetlock, “Investor sentiment and the influence of media on stock markets,” J. Finance, vol. 62, no. 3, pp. 1139–1168, 2007. https://doi.org/10.1111/j.1540-6261.2007.01232.x.
    16. T. O. Sprenger, A. Tumasjan, P. G. Sandner, and I. M. Welpe, “Infor- mation value of Twitter stock microblogs,” Eur. Financ. Manag., vol. 20, no. 5, pp. 926–957, 2014. https://doi.org/10.1111/j.1468-036X.2013.12007.x.
    17. H. Yu, J. Nartea, C. Gan, and L. J. Yao, “Profitability of technical trading rules in Southeast Asian markets,” Int. Rev. Econ. Finance, vol. 25, pp. 356–371, 2013. https://doi.org/10.1016/j.iref.2012.07.016.
    18. B. Weng, F. F. Ahmed, and F. M. Megahed, “Short-term stock movement prediction using heterogeneous data sources,” Expert Syst. Appl., vol. 79,pp. 153–163, 2017. https://doi.org/10.1016/j.eswa.2017.02.041.
    19. L. Kristoufek, “Drivers of Bitcoin price identified through wavelet coherence,” PLoS ONE, vol. 10, no. 4, e0123923, 2015. https://doi.org/10.1371/journal.pone.0123923.
    20. M. Ballings, D. Van den Poel, N. Hespeels, and R. Gryp, “Comparative evaluation of classifiers for predicting stock price direction,” Expert Syst. Appl., vol. 42, no. 20, pp. 7046–7056, 2015. https://doi.org/10.1016/j.eswa.2015.05.013.
    21. Z. Yang et al., “FinBERT-X: Domain-Specific Transformer Models for Financial Sentiment Analysis,” Expert Systems with Applications, vol. 213, 2023.
    22. L. Chen and J. Wang, “RoBERTa-Finance: Improved Contextual Representations for Financial NLP,” IEEE Access, vol. 11, pp. 55924–55937, 2023.

  • Downloads

  • How to Cite

    Kadu , M. L. ., Deshpande , D. M. M. ., & Pawar , D. V. N. . (2025). Market Trend Discovery Through Deep ‎Learning-Driven Twitter Sentiment for ‎Stock Forecasting. International Journal of Basic and Applied Sciences, 14(7), 388-395. https://doi.org/10.14419/c1qpgz81