Article: Supply-demand matching and personalised recommendation system of educational resources based on deep learning Journal: International Journal of Information and Communication Technology (IJICT) 2026 Vol.27 No.81 pp.48 - 70 Abstract: With the boom in educational informatisation and online learning platforms, the mismatch between educational resource supply and learners' personalised needs has become acute, hampering resource allocation efficiency and learning outcomes. This study proposes a deep learning-based educational resource recommendation system for demand-supply matching. First, we model educational resource features and learning needs, then design a deep learning matching model integrating multi-source features, and formulate personalised recommendation strategies using matching probabilities. Depending on the experimental verification, it can be found that the proposed method achieves a relatively stable performance in the task of supply and demand matching, and its F1-value reaches 0.829, which improves its performance by about 17% compared with the traditional collaborative filtering method. Ablation experiments confirm that incorporating learner and resource features significantly boosts model performance, and the method exhibits superior hit and resource completion rates, verifying its effectiveness in enhancing resource allocation efficiency. Inderscience Publishers - linking academia, business and industry through research

Title: Supply-demand matching and personalised recommendation system of educational resources based on deep learning

Authors: Mei Hao; Yiping Dai

Addresses: Zhizhen Academy, Henan Quality Institute, Pingdingshan, 467000, China ' School of Modern Information Technology, Henan Industry and Trade Vocational College, Zhengzhou, 451191, China

Abstract: With the boom in educational informatisation and online learning platforms, the mismatch between educational resource supply and learners' personalised needs has become acute, hampering resource allocation efficiency and learning outcomes. This study proposes a deep learning-based educational resource recommendation system for demand-supply matching. First, we model educational resource features and learning needs, then design a deep learning matching model integrating multi-source features, and formulate personalised recommendation strategies using matching probabilities. Depending on the experimental verification, it can be found that the proposed method achieves a relatively stable performance in the task of supply and demand matching, and its F1-value reaches 0.829, which improves its performance by about 17% compared with the traditional collaborative filtering method. Ablation experiments confirm that incorporating learner and resource features significantly boosts model performance, and the method exhibits superior hit and resource completion rates, verifying its effectiveness in enhancing resource allocation efficiency.

Keywords: deep learning; educational resources; supply and demand matching; personalised recommendation; recommendation system.

DOI: 10.1504/IJICT.2026.155113

International Journal of Information and Communication Technology, 2026 Vol.27 No.81, pp.48 - 70

Received: 11 Feb 2026
Accepted: 10 Apr 2026
Published online: 27 Jul 2026 *

Title: Supply-demand matching and personalised recommendation system of educational resources based on deep learning

Keep up-to-date