Article: Intelligent instructional resource management incorporating emotional and semantic features of user comments Journal: International Journal of Reasoning-based Intelligent Systems (IJRIS) 2025 Vol.17 No.10 pp.10 - 19 Abstract: The proliferation of digital teaching resources has exacerbated challenges in personalised recommendation due to information overload. This study introduces an intelligent management framework that integrates emotional and semantic features extracted from user-generated comments. By employing an adaptive weighting mechanism, multimodal feature fusion is achieved by analysing emotional intensity in user comments and dependencies among educational entities, utilising an adaptive weighting mechanism. Experimental evaluations on the EdNet public dataset reveal a 12.7% improvement in recommendation accuracy and a 9.2% increase in F1-score. These enhancements not only significantly optimise the assessment of resource quality but also improve the delivery of personalised services, thereby underscoring the framework's effectiveness in advancing educational resource management. Furthermore, this approach addresses critical limitations in existing systems and provides scalable solutions suitable for real-world applications. Inderscience Publishers - linking academia, business and industry through research

Title: Intelligent instructional resource management incorporating emotional and semantic features of user comments

Authors: Chenglin Lu

Addresses: Macau University of Science and Technology, Macau 999078, China

Abstract: The proliferation of digital teaching resources has exacerbated challenges in personalised recommendation due to information overload. This study introduces an intelligent management framework that integrates emotional and semantic features extracted from user-generated comments. By employing an adaptive weighting mechanism, multimodal feature fusion is achieved by analysing emotional intensity in user comments and dependencies among educational entities, utilising an adaptive weighting mechanism. Experimental evaluations on the EdNet public dataset reveal a 12.7% improvement in recommendation accuracy and a 9.2% increase in F1-score. These enhancements not only significantly optimise the assessment of resource quality but also improve the delivery of personalised services, thereby underscoring the framework's effectiveness in advancing educational resource management. Furthermore, this approach addresses critical limitations in existing systems and provides scalable solutions suitable for real-world applications.

Keywords: intelligent education; resource management; user comments mining; feature fusion; quality assessment.

DOI: 10.1504/IJRIS.2025.148175

International Journal of Reasoning-based Intelligent Systems, 2025 Vol.17 No.10, pp.10 - 19

Received: 15 Jun 2025
Accepted: 14 Jul 2025
Published online: 27 Aug 2025 *

Title: Intelligent instructional resource management incorporating emotional and semantic features of user comments

Keep up-to-date