Research on the Design and Effectiveness Analysis of Artificial Intelligence-Driven Intelligent Teaching and Assisting System for Civics and Political Science Courses

Abstract

The innovative development of the Civics and Political Science class against the backdrop of artificial intelligence represents a relatively new research direction within the academic world. This paper designs a teaching aid system for civics and political science classes based on artificial intelligence technology, with the main functions of accurate teaching intervention and personalized course resource recommendations. We incorporate the reinforcement learning algorithm into the decision-making process for accurate teaching interventions, build a decision-making model for these interventions based on reinforcement learning, and utilize the Q-learning algorithm to model the decision-making model’s data. Construct the Civics and Politics course resource network, update the keyword weights of the learning interest nodes, mine the interest features, and complete the personalized recommendation of the Civics and Politics course resources in colleges and universities based on collaborative filtering. We apply the system of this paper to the practice of teaching civics and politics to 50 first-year students majoring in ideology and politics at University D. After the practice, the students achieved an overall assessment score of 88.76, a 6.25 improvement over their pre-practice score, and demonstrated improved behavior and civics learning outcomes in the “excellent learners” and “potential learners” categories. The average satisfaction levels in the four aspects of system design, application effect, function evaluation, and function implementation effect are 4.625, 4.635, 4.627, and 4.647, respectively, with a high level of satisfaction.

Keywords

• artificial intelligence
• reinforcement learning algorithm
• q-learning algorithm
• collaborative filtering
• civics teaching
• 03b70