Work

Digital portrait and intelligent recommendation system for physical education teaching based on hybrid collaborative filtering

Year: 2026
Type: article
Abstract: This study aims to effectively address issues in traditional physical education (PE) teaching, such as blind course selection, mismatched teaching resources, and lack of personalized guidance. This study investigates the development of a digital portrait and intelligent recommendation system for physical education instruction based on hybrid collaborative filtering (HCF). The system constructs digital portraits of students for PE course scenarios by integrating multidimensional information—including basic attributes, physical abilities, learning behaviors, and interest preferences. It further uses adaptive feature weights to capture individual differences among students. In the recommendation model, a user interest confidence mechanism is introduced to construct user-side similarity relationships driven by digital portraits. Meanwhile, a course-side similarity relationship is established by incorporating a popularity correction strategy. Through the fusion of information from both sides, personalized recommendations for PE courses are achieved. The results demonstrate that the HCF-based digital portrait consruction and intelligent recommendation system for PE can stably accomplish course recommendation tasks. The experimental results demonstrate that the proposed model maintains strong recommendation performance on the public education dataset, the self-constructed PE dataset, and the external recommendation dataset. It also exhibits robust adaptability in cold-start user scenarios. Further teaching validation experiments indicate that the intelligent recommendation-based course matching approach can improve the suitability of students’ course selections. It also exerts positive effects on learning engagement and course performance. This study provides a feasible technical solution for personalized course recommendation and digital teaching management in PE, offering valuable insights for the integration of information technology into PE reform.
Language: English
Open Access status: gold
Estimated APC paid: $1,590
OpenAlex ID: W7235446626
FWCI:
Cites: 19
Cited by:
Related to: 10