An empirical study on students’ personalized teaching mode and educational effect of combining artificial intelligence technology in college civics education

Abstract

With the rapid popularization of the new generation of information technology, education informatization has become a major trend in today’s education development, and more and more intelligent educational assistance platforms have emerged. The study proposes an empirical Q-matrix neurocognitive diagnostic model by introducing a neurocognitive diagnostic model, which improves the diagnostic effect of the NCD-Q model by learning the complex interaction function between students and exercises through neural networks. CFR-CD, an enhanced personalized exercise recommendation technique, is proposed to create a learner model of similar exercise expertise using a collaborative filtering algorithm. The CFR-CD algorithm is tested experimentally against other recommendation algorithms by suggesting exercises that are based on students’ knowledge mastery and similar learners’ answer data. Among them, the recommended effect of the CFR-CD model on simple test questions is better than the NMF model, the precision rate of the CFR-CD model is improved by about 3% on 50% and 60% of test question test sets, and the recall rate and F1 value of this model are the best among the three models. Finally, the experimental pre and post-tests were conducted to explore the teaching effect of the CFR-CD model in Civics education, and three dimensions, as well as sub-dimensions, were divided to analyze the influence effect of the CFR-CD model, which concluded that the Civics character, cultural literacy, and value concepts were all improved, among which the improvement effect of Civics character was the most obvious.

Keywords

• civic education
• cognitive diagnosis
• personalized recommendation
• neural network
• assisted teaching
• 97m80