Research on Learning Path Recommendation Strategy Based on Knowledge Graph

Author(s)

Abstract
Personalized learning path recommendation is crucial for enhancing learning efficiency and engagement in increasingly diverse and large-scale online educational environments. Traditional recommendation methods often struggle to capture the complex semantic relationships and prerequisite dependencies inherent in learning domains. This paper proposes a novel learning path recommendation strategy based on Knowledge Graph. We construct a domain-specific educational KG integrating learning resources, concepts, skills, and their rich interrelationships (prerequisites, relatedness, difficulty levels). Based on this structured knowledge representation, we design a recommendation algorithm that utilizes graph traversal techniques and semantic similarity measures. The algorithm considers the learner's current knowledge state, target learning objectives, and individual preferences to dynamically generate optimal and coherent learning sequences. Experimental results demonstrate that our KG-based approach significantly outperforms baseline methods (e.g., collaborative filtering, content-based filtering) in terms of recommendation accuracy, path coherence, learning efficiency prediction, and learner satisfaction. This research provides a robust and explainable framework for personalized education, paving the way for more intelligent and adaptive learning systems.

Keywords
knowledge graph, personalized learning, learning path recommendation, graph algorithms, adaptive learning, semantic representation.

Cite this paper
Hongxia Wang, Research on Learning Path Recommendation Strategy Based on Knowledge Graph , SCIREA Journal of Computer. Volume 10, Issue 3, June 2025 | PP. 58-65. 10.54647/computer520465

References