Personalised Recommendation of Literary Learning Resources Based on a Mixed Recommendation of Learning Interest and Contextual Awareness

Abstract

In an effort to improve the efficiency and recommendation accuracy of mobile learning resources, the study proposes a hybrid mobile learning strategy based on Collaborative Filtering (CF), context and interest. Analyse from the perspective of situational awareness, construct a personalised recommendation model for text learning resources based on GimbalTM, and obtain a recommendation form. The experimental results show that the RMSE and MAE of Context-Collaborative filtering (C-CF) are lower than those of traditional CF. The Precision and Recall values of C-CF are higher than those of CF at 10 s, the recommendation growth rates of traditional CF and C-CF are 2.09% and 1.67%, respectively. The Gimbal software enables a certain degree of learner location detection and can trigger contextual rules based on time and location contexts to provide users with personalised text-based learning resources. The research results indicate that in specific applications, over time, under the recommendation system, students’ grades steadily increase, which is also beneficial for improving their learning efficiency.