From news to trends: a financial time series forecasting framework with LLM-driven news sentiment analysis and selective state spaces
Journal of Intelligent Information Systems · 2025-12 · 63 (6) · 1955-1980
DOI: 10.1007/s10844-025-00971-3 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Intelligent Information Systems, online erschienen am 02.08.2025. Verfasst von Renjie Wang und 2 weiteren Autor:innen. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 48 Literaturangaben. Gefördert von: Program of Science and Technol- ogy Development Plan of Jilin Province of China, Shenzhen Science and Technology Program.
Abstract (Originalsprache)
Stock price prediction is inherently challenging due to market volatility and the influence of external factors. Traditional forecasting methods primarily rely on historical price data, limiting their ability to capture market sentiment embedded in financial news. To address this limitation, we propose MambaMoE, a novel model that integrates historical stock prices with sentiment information extracted from financial news. Specifically, we fine-tune a DeepSeek-based large language model (LLM) for financial sentiment classification and incorporate the extracted sentiment information into our predictive framework. At the core of our approach is MambaMoE layer, which leverages the efficiency of state space models (SSMs) to model long-range dependencies while maintaining linear computational complexity, making it well-suited for financial time series forecasting. Additionally, the Mixture of Experts (MoE) mechanism improves the model’s ability to capture diverse market behaviors by dynamically selecting specialized experts based on stock data patterns. Experimental results demonstrate that MambaMoE outperforms LSTM-based models by 23.7% and Transformer-based models by 6.3%, highlighting its superior performance in short-term stock prediction.
Förderung
- Program of Science and Technol- ogy Development Plan of Jilin Province of China
- Shenzhen Science and Technology Program
Zitieren
APA
Wang, R., Sun, M., & Wang, L. (2025). From news to trends: a financial time series forecasting framework with LLM-driven news sentiment analysis and selective state spaces. Journal of Intelligent Information Systems, 63(6), 1955-1980. https://doi.org/10.1007/s10844-025-00971-3
Vancouver
Wang R, Sun M, Wang L. From news to trends: a financial time series forecasting framework with LLM-driven news sentiment analysis and selective state spaces. Journal of Intelligent Information Systems. 2025;63(6):1955-1980. doi:10.1007/s10844-025-00971-3
BibTeX
@article{Wang2025,
author = {Wang, Renjie and Sun, Minghui and Wang, Limin},
title = {From news to trends: a financial time series forecasting framework with LLM-driven news sentiment analysis and selective state spaces},
journal = {Journal of Intelligent Information Systems},
year = {2025},
volume = {63},
number = {6},
pages = {1955--1980},
doi = {10.1007/s10844-025-00971-3}
}
EUVIMED