Civic education reform based on deep reinforcement learning model

Abstract

Abstract The integration of artificial intelligence technology into education is an inevitable trend of scientific progress and educational reform, and how to use artificial intelligence technology and ideological and political education reform is called a key research direction in the education sector. Aiming at the problems of cold start in personalized recommendation system, lack of interpretability of recommendation results, and ignoring the implicit features of the course for better acceptance of recommendation results by learners, the BPRMF model based on deep learning is proposed to be applied to the problem of recommendation of Civics and Political Science course, which not only models learners’ preferences and combines with course attribute features to generate recommendation rating ranking list and provide personalized recommendation service. Then the study of Civics education reform is conducted, mainly analyzing the change in teaching methods based on big data, machine learning, and deep learning technologies to promote secondary school students. The performance of the BPRMF model is evaluated in comparison with the BPRMF model under different k values. It is concluded that the accuracy rate of the BPRMF model is 8.9%~12.01% higher than UBCF and 8.07%~10.26% higher than IBCF, but with the increase of k value, the recall rate will gradually pull away from other models and optimize the recommendation system to some extent. This study is beneficial to ideological education in the implementation process to better utilize the opportunities, meet the challenges, and develop efficiently.