Liu Jun
ABSTRACT
This study integrates the Random Forest algorithm and knowledge graph technology to construct a Students’ Learning Diagnosis and Evaluation System with Decision Support. The system supports students in online learning and testing, dynamically diagnosing their knowledge mastery and identifying weak links through learning trajectories and test data, while generating personalized learning paths using knowledge graphs. Teachers can optimize teaching strategies based on visualized learning analytics, and administrators obtain references for instructional decision-making through multi-source data integration. Empirical research, conducted with students from six classes of the same major at Nanchang Normal University of Applied Technology as the research subjects, shows through comparative experiments that the system significantly enhances students’ motivation for autonomous learning, helps learners accurately locate deficiencies, and promotes the achievement of personalized learning goals. Meanwhile, it notably improves the efficiency of teachers’ feedback on student learning, providing a data-driven scientific decision-making basis for educational management. This research provides a new pathway for personalized learning support and teaching decision optimization in smart education scenarios.
Keywords: Decision support, knowledge graph, learning diagnosis, learning evaluation, personalized learning
https://doi.org/10.57180/ltlw5317