Article: An attention-enhanced knowledge graph approach for personalised English vocabulary recommendation Journal: International Journal of Information and Communication Technology (IJICT) 2026 Vol.27 No.91 pp.70 - 90 Abstract: Personalised English vocabulary recommendation remains difficult because learner behaviour changes quickly while lexical relations are highly structured. To address this issue, this study proposes an attention-enhanced knowledge graph recommendation model for vocabulary learning. First, learner interaction sequences are encoded to capture recent review rhythm, repeated errors, and short-term preference shifts. Then, lexical relations are organised through a vocabulary knowledge graph to preserve semantic, derivational, and pedagogical connections among words. Finally, both sides are fused in a personalised scoring stage to generate the next-word recommendation list. Experimental results show that the proposed method achieved 0.742 Precision, 0.694 Recall, 0.781 NDCG, and 0.726 MRR on Dataset A, and remained consistently superior on Dataset B. The model shows strong ranking accuracy, stable recommendation behaviour, and good educational suitability. Inderscience Publishers - linking academia, business and industry through research

Title: An attention-enhanced knowledge graph approach for personalised English vocabulary recommendation

Authors: Kang Gao

Addresses: School of Foreign Studies, North China University of Water Resources and Electric Power, Zhengzhou, 450046, China

Abstract: Personalised English vocabulary recommendation remains difficult because learner behaviour changes quickly while lexical relations are highly structured. To address this issue, this study proposes an attention-enhanced knowledge graph recommendation model for vocabulary learning. First, learner interaction sequences are encoded to capture recent review rhythm, repeated errors, and short-term preference shifts. Then, lexical relations are organised through a vocabulary knowledge graph to preserve semantic, derivational, and pedagogical connections among words. Finally, both sides are fused in a personalised scoring stage to generate the next-word recommendation list. Experimental results show that the proposed method achieved 0.742 Precision, 0.694 Recall, 0.781 NDCG, and 0.726 MRR on Dataset A, and remained consistently superior on Dataset B. The model shows strong ranking accuracy, stable recommendation behaviour, and good educational suitability.

Keywords: personalised vocabulary recommendation; knowledge graph; attention mechanism; English learning.

DOI: 10.1504/IJICT.2026.155778

International Journal of Information and Communication Technology, 2026 Vol.27 No.91, pp.70 - 90

Received: 10 Apr 2026
Accepted: 18 May 2026
Published online: 13 Aug 2026 *

Title: An attention-enhanced knowledge graph approach for personalised English vocabulary recommendation

Keep up-to-date