Find in Library

The issue of strengthening the construction of economics courses, enhancing the moral development of economics professionals, and cultivating professionals with firm and lofty economics ideals is of utmost urgency. In this paper, we model the potential relationship prediction and global correspondence of economics curriculum civics in colleges and universities and then use the sequence markers specific to the subset of relationships to deal with the overlap between entities and entities to complete the ternary extraction work and to realize the joint curriculum civics element entity relationship extraction. Fusion of entity attributes and structural features can be achieved by utilizing the dual attention mechanism, and hyperbolic spatial embedding in graph structure feature extraction can be utilized for entity alignment. The economics program at a university in Guangzhou implements classroom civics integration teaching reform after completing the named entity recognition task test on the dataset. It was found that compared to LEBRT, the current state-of-the-art method for incorporating lexicon information, the model in this paper achieved a performance improvement of 2.53%, 2.26%, and 3.47% in the three metrics, respectively. After one semester of teaching reform and model application, the four scores of Economics Interest, Economics Habit, Economics Confidence, and Economics Autonomous Learning Ability increased by 28.13, 28.65, 18.39, and 26.68 points, respectively, compared with the pre-test, and professionalism also improved by more than 10 points in all cases. This paper provides a useful exploration for optimizing classroom civics teaching methods and efficiency, as well as combining implicit civics with cutting-edge information technology.