Exploring Cybersecurity Based on Educational Knowledge Graph Construction and Reasoning in Pre-trained Language Models

Abstract

In this paper, we describe a new framework for educational knowledge graph construction and reasoning that leverages pre-trained language models to enhance both educational content organization and network security through controlled information flow. The proposed approach combines advanced language processing understanding with structured knowledge representation in order to solve problems related to the organization and inference of complex educational knowledge. This paper reports new methodologies for knowledge graph construction and reasoning which involve both semantic and structural information within the educational domain. The framework demonstrated superior performance in comprehensive experiments, achieving significant improvements in knowledge extraction and reasoning tasks, when compared to prior art. Our methods achieved an F1-score of 0.853 in knowledge extraction and a path accuracy of 0.786 in reasoning tasks, outperforming the baseline methods by 8.2% and 8.5%, respectively. The practical use of the framework in the educational technology field indicates that it can be used to improve the effectiveness of personalized learning, content organization, and intelligent tutoring systems.