Research on the Generation and Recommendation Algorithm of Civics Teaching Content for Engineering Management Course Based on Knowledge Graph and Machine Learning

Curriculum Civics refers to the integration of Civics elements into the teaching of professional courses, so that courses other than Civics courses can also play the role of Civics teaching. In this paper, we study a knowledge mapping-based content generation technology for teaching course Civics and Politics, so that the knowledge of Civics and Politics courses can be integrated and visualized. The knowledge points, concepts, definitions and other information of the course Civics and Politics are extracted in the form of Civics and Politics knowledge triples. Through the extraction of the knowledge entity of curriculum Civics and politics, the relationship between semi-structured data and unstructured data is extracted to realize the integration of knowledge and content generation. After achieving content generation, the generated content is personalized through a deep reinforcement learning recommendation algorithm based on diversity optimization. Taking the two courses of Engineering Cost Management and Engineering Economics in the engineering management specialty as an example, it is found that the proposed knowledge graph construction method has an accuracy rate of 96.2%, which is able to effectively establish the knowledge association between the civic elements and the elements of professional knowledge, and realize the mining and generation of the civic elements. Meanwhile, the DDRL-Base recommendation algorithm achieves the optimum in accuracy, recall and F1 value indexes, and optimizes the problems such as cold start and sparse data in resource matrix, which improves the effect of recommending the Civics and Politics teaching content of the course.