Research on the Application of Talent Cultivation in Higher Education Based on Data Mining

In the era of big data and AI, higher education is adopting data-driven decision-making. This study presents a three-layer framework—“objectives-competency-behavior”—using sequence modeling and interpretable learning to uncover talent development patterns. By analyzing courses, activities, and employment, it proposes resource allocation optimization via a closed-loop feedback mechanism. Results show improved learner pathway mapping and effective academic risk prediction and personalized learning recommendations. The study also addresses data privacy and ethical governance, offering a streamlined methodology for data collection to deployment. It demonstrates that integrating data mining with education can shift decisions from experience-based to evidence-based, enhancing training accuracy and management efficiency.