Educational–industry knowledge graphs for curriculum and industry alignment: A case study in artificial Intelligence–driven digital image processing

This paper proposes the Educational–Industry Knowledge Graph (EIKG) as a unified semantic framework for aligning higher education curricula with evolving industry requirements under the paradigms of Education 5.0 and Industry 5.0. The approach integrates agile Scrum-based instructional processes with knowledge graph modeling to capture relationships among learning activities, competencies, tools, datasets, and industry practices. The EIKG is formally defined as a weighted directed property graph, enabling the representation of prerequisite dependencies, activity complexity, and learner-specific mastery states. A case study is conducted on a Digital Signal Processing (DSP) course augmented with artificial intelligence components, where students engage in three sprint cycles to develop a virtual laboratory. Empirical results demonstrate improvements in student mastery and increased engagement, measured through project workload and activity completion. The framework supports adaptive learning-path optimization using constrained shortest-path and MDP/POMDP formulations, enabling personalized instruction based on the Zone of Proximal Development. Furthermore, the integration of Scrum artifacts into the knowledge graph enables real-time analytics, curriculum gap identification, and enhanced alignment with industry practices. The findings indicate that EIKG provides a scalable, data-driven foundation for bridging the gap between academic instruction and labor market needs.