Dynamic Update Mechanism of the Knowledge Graph for AI Literacy Course

Abstract

The rapid evolution of artificial intelligence has far outpaced the traditional 3–5-year revision cycles of university curricula, resulting in outdated instructional materials and limiting students’ ability to understand emerging technologies. To address this challenge, this paper proposes a dynamic update mechanism for AI literacy course resources based on a knowledge graph. The mechanism integrates three core components: a technology-driven agile response system, an industry–academia collaborative injection mechanism, and a data-driven feedback and optimization loop. The agile response system employs frontier-technology tracking and multi-level content alerts to identify outdated knowledge in real time. The collaborative injection mechanism incorporates up-to-date industrial cases, datasets, and application scenarios, ensuring the practical relevance of teaching materials. The data-driven optimization mechanism leverages learning analytics to prioritize revisions, detect anomalies, and support rapid, atomic-level updates. By establishing an institutionalized workflow covering knowledge tracking, priority setting, content co-creation, review, publication, and effectiveness evaluation, the proposed framework transforms teaching resources from static repositories into dynamic knowledge services. This approach significantly enhances the timeliness, accuracy, and utilization efficiency of instructional materials while strengthening industry–education integration and improving students’ engineering competencies. The dynamic update mechanism provides a sustainable pathway for AI literacy education to adapt to fast-changing technological landscapes and to cultivate future-ready talent.