Short-Term Stock Market Trend Prediction Driven by Artificial Intelligence - A Comprehensive Model Based on Large-Scale Multi-Source Data

Abstract

Short-term stock market trend prediction plays an important role in financial analysis and investment decision-making, yet it remains a challenging task due to market volatility and complex influencing factors. From a business data analytics perspective, this study investigates an artificial intelligence-driven framework for short-term stock market trend prediction based on large-scale multi-source financial data. The proposed approach integrates historical market data and constructed analytical features to capture short-term market dynamics and generate directional trend signals. Rather than focusing on point price prediction, the framework adopts a classification-based strategy to support next-day trend assessment. Model performance is evaluated using publicly available financial market data, and the results demonstrate that the proposed framework is capable of providing stable and interpretable prediction outcomes. In addition, the study discusses the practical application of the proposed framework within real-world financial analysis workflows. The results indicate that artificial intelligence techniques, when combined with structured feature construction, can serve as effective auxiliary tools for short-term market analysis. This research contributes to the application-oriented exploration of artificial intelligence methods in financial data analysis and provides practical insights for business-oriented market prediction tasks.