EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Research on the Application of Multimodal Information Fusion in Financial Data Prediction

Qingying Zhi

Journal of Computer Science and Artificial Intelligence · 27.09.2025 · 4 (2) · 1-4

DOI: 10.54097/c6jmgg90 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Computer Science and Artificial Intelligence, online erschienen am 27.09.2025. Verfasst von Qingying Zhi. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 5 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The high complexity and uncertainty of financial markets make it difficult for predictive models that rely on a single data source to effectively address these challenges. Multimodal information fusion, by integrating heterogeneous data sources such as text, numerical data, images, and time series, reveals hidden connections and opens new avenues for improving the accuracy and robustness of financial forecasts. This paper systematically analyzes the theoretical foundations, key technical approaches, and typical scenarios for applying multimodal information fusion to financial forecasting. The study first analyzes the characteristics and categories of multimodal financial data, encompassing structured market data, unstructured text, time series data streams, and visual information. Secondly, it focuses on the core technologies of multimodal fusion, covering the data layer, feature layer, decision layer, and hybrid fusion strategies. It also analyzes the advantages of deep learning models in processing and fusing heterogeneous data. Through specific application cases such as stock price trend forecasting, credit risk assessment, and macroeconomic indicator forecasting, this paper demonstrates the significant effectiveness of multimodal fusion in capturing market sentiment, identifying potential risks, and improving forecasting accuracy. The study also identifies key challenges currently faced, including data heterogeneity, model interpretability, computational efficiency, and privacy protection. Finally, we outline future research directions, emphasizing the importance of cross-modal alignment, adaptive fusion mechanisms, interpretability enhancement, and privacy-preserving fusion frameworks. This study provides theoretical support and practical references for deepening the application of multimodal fusion in intelligent financial decision-making.

Zitieren

APA

Zhi, Q. (2025). Research on the Application of Multimodal Information Fusion in Financial Data Prediction. Journal of Computer Science and Artificial Intelligence, 4(2), 1-4. https://doi.org/10.54097/c6jmgg90

Vancouver

Zhi Q. Research on the Application of Multimodal Information Fusion in Financial Data Prediction. Journal of Computer Science and Artificial Intelligence. 2025;4(2):1-4. doi:10.54097/c6jmgg90

BibTeX
@article{Zhi2025,
  author = {Zhi, Qingying},
  title = {Research on the Application of Multimodal Information Fusion in Financial Data Prediction},
  journal = {Journal of Computer Science and Artificial Intelligence},
  year = {2025},
  volume = {4},
  number = {2},
  pages = {1--4},
  doi = {10.54097/c6jmgg90}
}
Kostenlos starten →