PERSONALIZED LEARNING PATHS FOR MENTAL HEALTH EDUCATION USING KNOWLEDGE GRAPH AND NLP

Abstract Objective Enhancing mental health literacy requires innovative educational approaches. This study aimed to develop and assess a framework for automatically generating personalized learning paths in mental health education, with the goal of improving learning efficiency through customized content sequences tailored to individual learners. Methods We created a system based on Knowledge Graph (KG) and Natural Language Processing (NLP). A mental health KG was built, representing key concepts and their relationships. NLP processed learner inputs to identify starting points. In an evaluation with 120 participants, the experimental group (n = 60) used the KG-NLP system, while the control group (n = 60) followed a standard, linear curriculum. The system generated unique paths by navigating the KG according to each user’s profile and progress. Results The experimental group demonstrated significantly better outcomes, achieving scores 28% higher on a post-test than the control group (p<.01). User satisfaction was markedly higher, with 92% of users rating the recommended paths as “highly relevant.” The system effectively adapted to varying knowledge levels, personalizing the learning journey for each individual. Conclusions The integration of KG and NLP offers an effective method for personalizing mental health education. This approach facilitates more efficient knowledge acquisition and increases learner engagement, demonstrating significant potential for developing adaptable and effective public health education tools. Acknowledgment The research was supported by: 2019 Guangdong Province Characteristic Innovation Project “Research on Online Learning Demonstration System Based on Knowledge Graph-Taking Data Structure Courses as an Example” (Project No.: 2019KTSCX225).