DGRCL:A Dynamic Contrastive Graph Learning Framework for Modeling Stock Market Evolution

Key Points

• The dynamic contrastive graph learning framework improves modeling of stock market evolution, focusing on predictive accuracy.
• Key evidence includes observing significant enhancements in prediction capabilities, especially during volatile market phases.
• This analysis employs a novel graph-based framework for data representation and interaction within stock market datasets.
• These findings may enable better forecasting methods in finance, indicating potential for real-world applications in trading.