Author(s)

Hailong Sun ¹, Adisak Sangsongfa ¹, Noppadol Amdee ¹

Affiliation(s)

¹ Faculty of Industrial Technology, Muban Chom Bueng Rajabhat University, 70150, Ratchaburi, Thailand

Corresponding Author

Adisak Sangsongfa

ABSTRACT

Stock market trend prediction remains a technically demanding task, partly because price movements are shaped by both quantitative trading patterns and harder-to-quantify factors such as investor sentiment. Most existing methods address these two aspects in isolation, which may limit their robustness under volatile or news-driven market conditions. This paper describes the development and preliminary evaluation of T-VAETrans, a hybrid architecture that combines a Variational Autoencoder (VAE) with a Transformer encoder to model sentiment-laden and sequential financial data within a single framework. The VAE component learns probabilistic latent representations of sentiment features collected from Chinese financial news and social platforms, while the Transformer handles temporal modeling of technical indicators via multi-head self-attention. A gated fusion layer adaptively weights the contributions of these two streams based on inferred market context. Experiments on a five-year dataset of 50 stocks from the Shanghai Stock Exchange (SSE) show that the proposed approach achieves an overall classification accuracy of 87.3%, compared to 81.4% for the best-performing baseline (TRA-LSTM). Ablation results confirm that both the VAE uncertainty module and the Transformer temporal module contribute meaningfully to the observed performance. Several practical limitations, including dependence on Chinese-language sentiment sources and the computational overhead of the hybrid architecture, are discussed alongside directions for further investigation.

KEYWORDS

Stock Market Prediction; Sentiment Analysis; Variational Autoencoder; Transformer; Multi-Modal Deep Learning; Shanghai Stock Exchange

CITE THIS PAPER

Hailong Sun, Adisak Sangsongfa, Noppadol Amdee. Sentiment-Driven Stock Market Forecasting with a Hybrid Deep Learning Framework. Journal of Electronics and Information Science (2026). Vol. 11, No. 1, 42-51. DOI: http://dx.doi.org/10.23977/10.23977/jeis.2026.110106.

REFERENCES

[1] Jing N, Wu Z, Wang H. A hybrid model integrating deep learning with investor sentiment analysis for stock price prediction. Expert Systems with Applications, 2021, 178: 115019. https://doi.org/10.1016/j.eswa.2021.115019
[2] Vaswani A, Shazeer N, Parmar N, et al. Attention is all you need. Advances in Neural Information Processing Systems, 2017, 30: 5998-6008. https://papers.nips.cc/paper/7181-attention-is-all-you-need
[3] Du H, Lv L, Wang H, et al. TGNS: A transformer-based graph neural network for stock trend forecasting. Information Sciences, 2025, 720: 122555. https://doi.org/10.1016/j.ins.2025.122555
[4] Zhang P, Harris R D F, Zheng J. GNN-based social media sentiment analysis for stock market forecasting and trading. Expert Systems with Applications, 2025, 291: 128425. https://doi.org/10.1016/j.eswa.2025.128425
[5] Zhen K, Xie D, Hu X. A multi-feature selection fused with investor sentiment for stock price prediction. Expert Systems with Applications, 2025, 278: 127381. https://doi.org/10.1016/j.eswa.2025.127381
[6] Gunduz H. An efficient stock market prediction model using hybrid feature reduction method based on variational autoencoders and recursive feature elimination. Financial Innovation, 2021, 7(1): 28. https://doi.org/10.1186/s40854-021-00243-3
[7] Abdullah M, Sulong Z, Chowdhury M A F. Explainable deep learning model for stock price forecasting using textual analysis. Expert Systems with Applications, 2024, 249(C): 123740. https://doi.org/10.1016/j.eswa.2024.123740

Sponsors, Associates, and Links

---

• Information Systems and Signal Processing Journal
  
  
• Intelligent Robots and Systems
  
  
• Journal of Image, Video and Signals
  
  
• Transactions on Real-Time and Embedded Systems
  
  
• Journal of Electromagnetic Interference and Compatibility
  
  
• Acoustics, Speech and Signal Processing
  
  
• Journal of Power Electronics, Machines and Drives
  
  
• Journal of Electro Optics and Lasers
  
  
• Journal of Integrated Circuits Design and Test
  
  
• Journal of Ultrasonics
  
  
• Antennas and Propagation
  
  
• Optical Communications
  
  
• Solid-State Circuits and Systems-on-a-Chip
  
  
• Field-Programmable Gate Arrays
  
  
• Vehicular Electronics and Safety
  
  
• Optical Fiber Sensor and Communication
  
  
• Journal of Low Power Electronics and Design
  
  
• Infrared and Millimeter Wave
  
  
• Detection Technology and Automation Equipment
  
  
• Journal of Radio and Wireless
  
  
• Journal of Microwave and Terahertz Engineering
  
  
• Journal of Communication, Control and Computing
  
  
• International Journal of Surveying and Mapping
  
  
• Information Retrieval, Systems and Services
  
  
• Journal of Biometrics, Identity and Security
  
  
• Journal of Avionics, Radar and Sonar