Article: Knowledge graph-based analysis and intelligent recommendation of entrepreneurship course content Journal: International Journal of Information and Communication Technology (IJICT) 2025 Vol.26 No.46 pp.62 - 75 Abstract: This research presents a knowledge graph-driven recommendation system for entrepreneurship courses, incorporating data collection, pre-processing, entity recognition, graph analysis, and recommendation algorithms. By combining semantic relationships with collaborative filtering, the system enhances personalisation, improving both accuracy and user satisfaction. Knowledge graphs provide a structured representation of entities and their relationships, enabling more relevant and context-aware recommendations. Leveraging this capability, the system aligns course suggestions with learners preferences and educational goals. Prior studies highlight the effectiveness of knowledge graphs in domains such as tourism, education, and e-commerce for improving recommendation precision. The proposed workflow follows sequential stages, including data pre-processing, knowledge extraction, graph construction, analysis, and algorithm development. integrating top-down ontology design with bottom-up entity extraction, guides knowledge graph creation. Experiments on a dataset of over 6,000 educational resources achieved stable accuracy after 80 training epochs using the BERT-BiGRU-MHSA-CRF framework. User evaluations confirmed higher engagement and satisfaction compared to baseline models. Inderscience Publishers - linking academia, business and industry through research

Title: Knowledge graph-based analysis and intelligent recommendation of entrepreneurship course content

Authors: Yeting Chen; Wenchi Chen

Addresses: College of Civil and Construction Engineering, Hunan Institute of Technology, Hengyang, Hunan, 431002, China ' School of Cyberspace Security, Hunan College of Information, Changsha, Hunan, 410200, China

Abstract: This research presents a knowledge graph-driven recommendation system for entrepreneurship courses, incorporating data collection, pre-processing, entity recognition, graph analysis, and recommendation algorithms. By combining semantic relationships with collaborative filtering, the system enhances personalisation, improving both accuracy and user satisfaction. Knowledge graphs provide a structured representation of entities and their relationships, enabling more relevant and context-aware recommendations. Leveraging this capability, the system aligns course suggestions with learners preferences and educational goals. Prior studies highlight the effectiveness of knowledge graphs in domains such as tourism, education, and e-commerce for improving recommendation precision. The proposed workflow follows sequential stages, including data pre-processing, knowledge extraction, graph construction, analysis, and algorithm development. integrating top-down ontology design with bottom-up entity extraction, guides knowledge graph creation. Experiments on a dataset of over 6,000 educational resources achieved stable accuracy after 80 training epochs using the BERT-BiGRU-MHSA-CRF framework. User evaluations confirmed higher engagement and satisfaction compared to baseline models.

Keywords: knowledge graph; recommendation system; entrepreneurship education; data pre-processing; semantic analysis; collaborative filtering.

DOI: 10.1504/IJICT.2025.150603

International Journal of Information and Communication Technology, 2025 Vol.26 No.46, pp.62 - 75

Received: 13 Aug 2025
Accepted: 19 Sep 2025
Published online: 17 Dec 2025 *

Title: Knowledge graph-based analysis and intelligent recommendation of entrepreneurship course content

Keep up-to-date