Answers are grounded in this paper and cite their sources.
To enhance the accuracy and stability of stock market pricing predictions, this study investigates the application of various machine learning algorithms empowered by financial technology in stock data analysis and constructs an optimized model based on ensemble learning. The study first reviews traditional stock pricing models, highlighting their limitations in addressing complex market environments. It then introduces novel machine learning models such as Transformer, temporal fusion Transformer, and graph neural network, leveraging ensemble learning to improve model performance. In the experimental design, eight specific evaluation metrics are established according to two dimensions: stock pricing prediction accuracy and market environment adaptability. A comprehensive performance comparison is conducted among four models.
No takes yet. Share an insight, caveat, or question.
Chen et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: