A method for Analyzing and Predicting Students' Competitive Performance via Temporal Knowledge Graph

This paper proposes a dynamic model for analyzing and predicting students' competitive performance using a Temporal Knowledge Graph (TKG). By integrating historical competition data and temporal factors, the model constructs a time-aware knowledge graph to quantify question difficulty, knowledge point weights, and the evolution of students' problem-solving abilities. Unlike traditional static evaluation methods, the TKG dynamically tracks students' mastery of algorithmic concepts over time, incorporating principles of the forgetting curve and time-decay weighting to analyze learning efficiency and skill development trends. This enables accurate predictions of future competition performance. Experimental results demonstrate that the model provides actionable insights for personalized education, allowing educators to design targeted training programs and optimize pedagogical strategies. The key advantage of the TKG lies in its temporal awareness and adaptability, offering a more precise reflection of students' evolving capabilities and serving as a novel analytical tool for competitive education.