Research on knowledge graph-driven adaptive learning path generation for vehicle and transportation engineering courses

Le résumé fourni par la source

To address the pedagogical problems that arise from the high coupling of knowledge systems and complex interdisciplinary logic in the fields of Vehicle and Transportation Engineering, this paper puts forward an adaptive learning path generation model based on knowledge graphs. First, by using multi-source heterogeneous data such as course textbooks, academic and career advancement specifications, vehicle-to-infrastructure (V2X) experimental data, and industry standards, ontology modelling and Natural Language Processing (NLP) techniques are employed for knowledge extraction. Identify the core engineering entities and their semantic correlations to build a domain-specific cognitive map of the intersection of Vehicle and Transportation Engineering, and then develop an improved Ant Colony Optimization (ACO) algorithm based on the topological structure of knowledge graphs for the generation of adaptive learning trajectories. Quantify the strength of the logical dependency between knowledge points, integrate multi-dimensional learner cognitive state data with an optimal learning gain mechanism, and thus achieve dynamic planning and precise delivery of learning paths. Based on the above empirical results, the new model can reduce the problem of information overload caused by the scattering of knowledge and improve students’ retrieval efficiency and deep-seated, systematic understanding of complex engineering systems.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.