Scholarlink

<p>The cultivation of cultural communication talents is crucial in preserving and promoting cultural identity in the digital era. Building a structured knowledge system and an optimized development path is essential for addressing the evolving demands of interdisciplinary communication and media environments. However, traditional educational models often suffer from fragmented knowledge delivery, limited adaptability, and a lack of personalization in learning trajectories. To address these limitations, this paper proposes a framework, Graph Neural Network with Hierarchical Node Classification (GNN-HNC), that models the complex interrelations among cultural knowledge, skills, and competencies. The proposed method constructs a knowledge graph integrating academic curricula, cultural domains, and industry skill requirements, where GNN-HNC is applied to classify learning nodes at multiple levels, basic, intermediate, and advanced, while dynamically predicting optimal learning paths. This system enhances personalized learning recommendations and curriculum design by capturing hierarchical dependencies and latent connections in cultural learning structures. Results show the GNN-HNC achieving 90% classification accuracy, a 23-point gain in communication skills, 83–88% adaptability across cultural clusters, and sustained engagement growth.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Construction of a knowledge system and optimization of the cultivation path of cultural communication talents based on graph neural network

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年8月4日

查看全文
  • Wei Huang,
  • Xueqing Liu,
  • Jing Zeng

摘要

The cultivation of cultural communication talents is crucial in preserving and promoting cultural identity in the digital era. Building a structured knowledge system and an optimized development path is essential for addressing the evolving demands of interdisciplinary communication and media environments. However, traditional educational models often suffer from fragmented knowledge delivery, limited adaptability, and a lack of personalization in learning trajectories. To address these limitations, this paper proposes a framework, Graph Neural Network with Hierarchical Node Classification (GNN-HNC), that models the complex interrelations among cultural knowledge, skills, and competencies. The proposed method constructs a knowledge graph integrating academic curricula, cultural domains, and industry skill requirements, where GNN-HNC is applied to classify learning nodes at multiple levels, basic, intermediate, and advanced, while dynamically predicting optimal learning paths. This system enhances personalized learning recommendations and curriculum design by capturing hierarchical dependencies and latent connections in cultural learning structures. Results show the GNN-HNC achieving 90% classification accuracy, a 23-point gain in communication skills, 83–88% adaptability across cultural clusters, and sustained engagement growth.