EdgeKG-EN: A Dynamic English Knowledge Graph Framework With Edge Computing-Driven Optimization

Abstract

Addressing the limitations of traditional cloud architectures in timeliness, heterogeneous adaptability, and energy efficiency, this paper presents EdgeKG-EN, an edge-intelligence-driven dynamic knowledge graph framework for adaptive English education. The framework establishes three core mechanisms: temporal attention-based dynamic graph modeling for real-time concept evolution tracking, lightweight knowledge distillation protocols that enable efficient edge-device updates, and reinforcement learning-based scheduling strategies that optimize resource allocation. Multimodal learning alignment ensures cognitive-semantic consistency while privacy-preserving mechanisms guarantee data security. Experiments demonstrate that the framework significantly enhances knowledge reasoning timeliness and personalized recommendation accuracy under low-power operation, providing a novel solution for distributed educational scenarios.