Intelligent Construction of Civic Teaching Resources for Ancient Literature Course Based on Natural Language Processing Technology

This paper completes the intelligent construction of Civics teaching resources for ancient literature course based on natural language processing technology. The automatic filtering of the elements of the Civics and Politics course is realized by using the TF-IDF statistical method and cosine similarity, combined with the improved BERT model. Based on the knowledge graph and the introduction of collaborative filtering algorithms, we construct a recommendation method for the teaching resources of the Civics and Politics of Ancient Literature course. The experimental results show that the combination of TF-IDF and improved BERT model in this paper can effectively complete the comprehensive scoring and automatic screening of the elements of Civics and Politics, and provide high-quality ancient literature course content for the recommendation of Civics and Politics teaching resources. The recommendation method based on knowledge graph outperforms other algorithms in resource coverage, P/N measure, and AUC measure, and is capable of comprehensively and accurately recommending ancient literature course Civics teaching resources to students.