Multimodal Deep Learning Ensemble Framework for Accurate Stock Market Prediction Using Multisource Data
DOI:
https://doi.org/10.22399/ijcesen.2578Keywords:
Multimodal Data Integration, Hybrid Ensemble Model, MDSFE Framework, Predictive Accuracy, Benchmarking, LSTMAbstract
Stock market forecasting presents substantial challenges due to the inherent volatility of financial data, impacted by a number of variables, including as investor sentiment and economic indices. This study proposes an advanced hybrid ensemble framework, MDSFE (Multimodal Deep Stock Forecasting Ensemble), which integrates multiple deep learning architectures such as LSTM, VMD-BiLSTM-AM, and RoBERTa-TextCNN. Utilizing a multimodal data assimilation strategy, MDSFE leverages historical stock prices, real-time financial news, social media sentiment, and economic indicators. Benchmarked against traditional models like ARIMA and standalone LSTM models, MDSFE demonstrates superior predictive accuracy, achieving an R² value of 0.97 and a MAPE of 0.80%. Trained on a robust dataset comprising 10,000 instances collected from 2003 to 2024, MDSFE highlights its practical applicability in real-world scenarios, offering enhanced decision-making capabilities for investors and analysts.
References
[1] Zhao, Y., & Zhang, H. (2020). Enhancing stock price prediction via market sentiment and economic indicators. Journal of Finance Markets. 48;100548. https://doi.org/10.1016/j.finmar.2020.100548
[2] Zhang, Y., Wu, J., & Li, Q. (2021). A hybrid model for stock price prediction using multimodal data. Expert Systems with Applications. 169;114284. https://doi.org/10.1016/j.eswa.2021.114284
[3] Adebayo, J. A., & Adetunji, A. (2021). Stock price prediction using BiLSTM neural network. Journal of Computational Science. 48;101217. https://doi.org/10.1016/j.jocs.2021.101217
[4] Devlin, J., Chang, M. W., Lee, K., & Toutanova, K. (2019). BERT: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint. arXiv:1810.04805
[5] Liu, Y., Ott, P. E., Goyal, N., Du, J., & McCann, B. (2019). RoBERTa: A robustly optimized BERT pretraining approach. arXiv preprint. arXiv:1907.11692
[6] Wu, H., & Xu, Y. (2021). Ensemble learning for stock price prediction: A comprehensive survey. Expert Systems with Applications. 174;114665. https://doi.org/10.1016/j.eswa.2021.114665
[7] Cheng, X., Li, Y., & Wu, J. (2022). Stock price prediction using a deep ensemble learning model. Journal of Economic Dynamics and Control. 133;104247. https://doi.org/10.1016/j.jedc.2022.104247
[8] Livieris, I., Pintelas, E., & Pintelas, A. (2020). A CNN–LSTM model for gold price time-series forecasting. Neural Computing and Applications. https://doi.org/10.1007/s00521-020-04867x
[9] Yan, X., Weihan, W., & Chang, M. (2020). Research on financial assets transaction prediction model based on LSTM neural network. Neural Computing and Applications. https://doi.org/10.1007/s00521-020-049927
[10] Wang, S. F., Chen, H. S., & Lin, T. Y. (2020). Deep learning for stock market prediction using technical indicators and financial news articles. IEEE Access. 8;300032-300047. https://doi.org/10.1109/ACCESS.2020.3000327
[11] Lin, Z., He, Y., & Zhang, X. (2021). Predicting stock market trends using LSTM networks with sentiment analysis. Journal of Forecasting. 40(8);1351-1365. https://doi.org/10.1002/for.2749
[12] Li, J., Zhao, L., & Wang, H. (2022). Multimodal deep learning model for stock price forecasting. Neural Computing and Applications. https://doi.org/10.1007/s00521-022-06829-x
[13] Chen, J., Lin, Z., & Sun, Y. (2021). Hybrid deep learning model for stock price forecasting using sentiment analysis. Neural Computing and Applications. https://doi.org/10.1007/s00521-021-05948-6
[14] Gupta, A., Sharma, R., & Patel, K. (2020). Predicting stock trends with deep learning: A comparative study of CNN and LSTM models. IEEE Access. 8;203432-203447. https://doi.org/10.1109/ACCESS.2020.3035462
[15] Huang, W., Nakamura, Y., & Zhou, X. (2022). A multimodal deep ensemble model for real-time stock market prediction. Journal of Computational Finance. 26(3);89-110. https://doi.org/10.1016/j.jcf.2022.07.005
[16] Singh, H., Verma, P., & Kumar, A. (2021). Attention-based BiLSTM model for enhanced stock market forecasting. IEEE Transactions on Knowledge and Data Engineering. 33(11);3157-3170. https://doi.org/10.1109/TKDE.2021.3062457
[17] Zhou, X., Wang, Y., & Chen, H. (2021). Stock market forecasting using transformer-based models with financial sentiment analysis. IEEE Transactions on Knowledge and Data Engineering. 34(5);2103-2117. https://doi.org/10.1109/TKDE.2021.3095637
[18] Patel, J., Shah, S., & Thakkar, P. (2020). Integrating deep learning and reinforcement learning for financial time series prediction. Expert Systems with Applications. 164;113918. https://doi.org/10.1016/j.eswa.2020.113918
[19] Wang, P., Liu, J., & Zhang, T. (2021). Multi-source data fusion for stock market prediction using deep neural networks. IEEE Access. 9;71345-71357. https://doi.org/10.1109/ACCESS.2021.3086359
[20] Kim, H., Lee, Y., & Choi, W. (2022). Analyzing stock market movements using hybrid deep learning frameworks. Journal of Financial Data Science. 8(2);75-89. https://doi.org/10.1016/j.jfds.2022.04.006
[21] Jiang, B., Chen, R., & Song, L. (2020). A novel multimodal approach for stock price prediction integrating text and numerical data. Neural Computing and Applications. https://doi.org/10.1007/s00521-020-05132-2
[22] Xu, Z., Li, C., & Zhang, K. (2021). Evaluating the impact of financial news sentiment on stock price movement with deep learning. IEEE Access. 9;56789-56802. https://doi.org/10.1109/ACCESS.2021.3074789
[23] Luo, W., Wang, Y., & Sun, M. (2022). Temporal attention-based LSTM model for high-frequency stock trading strategies. IEEE Transactions on Neural Networks and Learning Systems. 33(12);7856-7868. https://doi.org/10.1109/TNNLS.2022.3154805
[24] Venkatesh, S., Bhardwaj, A., & Srinivasan, P. (2020). A comparative study of machine learning and deep learning models for stock market forecasting. Neural Computing and Applications. https://doi.org/10.1007/s00521-020-05208-x
[25] Han, J., Wu, K., & Gao, Y. (2021). Enhancing stock trend prediction using an ensemble deep learning framework. IEEE Access. 9;89347-89359. https://doi.org/10.1109/ACCESS.2021.3090589
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 International Journal of Computational and Experimental Science and Engineering
This work is licensed under a Creative Commons Attribution 4.0 International License.
