Article: Research on the construction of higher vocational curriculum system and path optimisation model based on knowledge graph Journal: International Journal of Information and Communication Technology (IJICT) 2026 Vol.27 No.77 pp.57 - 76 Abstract: The lack of personalised adaptation in higher vocational curriculum systems is increasingly prominent. To address this, we propose a knowledge graph-based curriculum system construction and path optimisation model (KG-VCSPOM). The model integrates two modules: a knowledge graph construction module using BERT and GGAT to extract course entities and relationships, and a path optimisation module combining GCN embedding and reinforcement learning to compute optimal learning sequences. Tests show the model achieves an F1-score of 84.6% (baseline CF: 71.2%), a noise robustness attenuation rate of 6.2%, and a user satisfaction score of 4.5. Performance remains stable as data scales to 100,000 entries. These results confirm the model's advantages in improving recommendation accuracy, adaptability, and practicality, providing an innovative solution for intelligent higher vocational education. Inderscience Publishers - linking academia, business and industry through research

Title: Research on the construction of higher vocational curriculum system and path optimisation model based on knowledge graph

Authors: Wanli Li

Addresses: Chongqing Business Vocational College, Chongqing, China; Faculty of Arts, Communication and Education, Kuala Lumpur University of Science and Technology, Kuala Lumpur, Malaysia

Abstract: The lack of personalised adaptation in higher vocational curriculum systems is increasingly prominent. To address this, we propose a knowledge graph-based curriculum system construction and path optimisation model (KG-VCSPOM). The model integrates two modules: a knowledge graph construction module using BERT and GGAT to extract course entities and relationships, and a path optimisation module combining GCN embedding and reinforcement learning to compute optimal learning sequences. Tests show the model achieves an F1-score of 84.6% (baseline CF: 71.2%), a noise robustness attenuation rate of 6.2%, and a user satisfaction score of 4.5. Performance remains stable as data scales to 100,000 entries. These results confirm the model's advantages in improving recommendation accuracy, adaptability, and practicality, providing an innovative solution for intelligent higher vocational education.

Keywords: knowledge graph; higher vocational education; curriculum system; path optimisation.

DOI: 10.1504/IJICT.2026.154956

International Journal of Information and Communication Technology, 2026 Vol.27 No.77, pp.57 - 76

Received: 25 Feb 2026
Accepted: 29 Apr 2026
Published online: 20 Jul 2026 *

Title: Research on the construction of higher vocational curriculum system and path optimisation model based on knowledge graph

Keep up-to-date