Personalized Accurate Recommendation Algorithm of Ideological and Political Teaching Multimedia Resources Based on Mobile Learning

Abstract

The currently used resource recommendation algorithm mainly recommends resources according to the user's preference for a tag class, ignoring the relationship between user preferences and needs and learning scenarios under mobile learning, resulting in poor efficiency and accuracy of recommended resources. In order to improve the shortcomings of the algorithm, this paper studies the personalized recommendation algorithm of Ideological and political teaching multimedia resources based on mobile learning. By constructing the map of Ideological and political teaching knowledge, this paper analyzes the correlation between resources. The diagnosis result of students' cognitive level is one of the characteristics of personalized recommendation. Mobile learning devices are used to collect data, calculate and perceive mobile learning scenarios. By improving the collaborative filtering technology, the teaching resources of Ideological and political courses can be personalized recommended. In the algorithm experiment, the average absolute error of the algorithm recommendation is relatively reduced by about 14.67%, the recommendation efficiency is higher, and the personalized recommendation effect is better.