Article: Deep interest fusion for cross-modal recommendation of English teaching resources Journal: International Journal of Computer Applications in Technology (IJCAT) 2026 Vol.78 No.6 pp.68 - 77 Abstract: In order to fill the theoretical gap in the field of personalised matching of multimodal learning resources, a cross-modal recommendation method for English teaching resources based on deep fusion of interest information is studied. Firstly, collect and analyse user behaviour data, attribute features and learning characteristics on the English learning platform, and construct a multidimensional interest fusion model. Secondly, by integrating the BERT model with cross-modal attention perception methods, a CCA-BERT recommendation model was constructed to achieve deep feature extraction and semantic association modelling of multimodal English teaching resources such as videos, texts and audios. Finally, personalised resource recommendation is completed based on click probability, which breaks through the limitations of traditional single mode recommendation. Empirical findings demonstrate that our cross-modal recommendation approach achieves a user satisfaction level exceeding 94.7%, while simultaneously maintaining recommendation diversity above 0.87% across experimental evaluations. Inderscience Publishers - linking academia, business and industry through research

Title: Deep interest fusion for cross-modal recommendation of English teaching resources

Authors: Jinghui Liu; Jin Zhang; Xue Cao

Addresses: School of Foreign Languages, Handan University, Handan, Hebei, China ' School of Foreign Languages, Handan University, Handan, Hebei, China ' School of Foreign Languages, Handan University, Handan, Hebei, China

Abstract: In order to fill the theoretical gap in the field of personalised matching of multimodal learning resources, a cross-modal recommendation method for English teaching resources based on deep fusion of interest information is studied. Firstly, collect and analyse user behaviour data, attribute features and learning characteristics on the English learning platform, and construct a multidimensional interest fusion model. Secondly, by integrating the BERT model with cross-modal attention perception methods, a CCA-BERT recommendation model was constructed to achieve deep feature extraction and semantic association modelling of multimodal English teaching resources such as videos, texts and audios. Finally, personalised resource recommendation is completed based on click probability, which breaks through the limitations of traditional single mode recommendation. Empirical findings demonstrate that our cross-modal recommendation approach achieves a user satisfaction level exceeding 94.7%, while simultaneously maintaining recommendation diversity above 0.87% across experimental evaluations.

Keywords: interest information; information mining; English teaching resources; cross-modal recommendation.

DOI: 10.1504/IJCAT.2026.153741

International Journal of Computer Applications in Technology, 2026 Vol.78 No.6, pp.68 - 77

Received: 30 Sep 2025
Accepted: 09 Dec 2025
Published online: 22 May 2026 *

Title: Deep interest fusion for cross-modal recommendation of English teaching resources

Keep up-to-date