现代情报2026，Vol.46Issue(2)：172-184,13.DOI:10.3969/j.issn.1008-0821.2026.02.015
基于表示学习的跨学科概念关联研究
Research on Interdisciplinary Conceptual Knowledge Fusion Based on Representation Learning
摘要
Abstract
[Purpose/Significance]To address the challenges in deep representation learning of concepts within multi-level semantic relationships and their interdisciplinary associations,and to break through the limitations of traditional me-thods that rely solely on surface feature matching.[Method/Process]This paper proposes an interdisciplinary concept associa-tion method based on a subject concept knowledge graph.This method leverages a representation learning-based know-ledge alignment model to integrate syntactic,semantic,and pragmatic correlations,capturing the implicit structural asso-ciation features within the subject knowledge graph to construct a concept association model oriented towards interdiscipli-nary knowledge services.[Result/Conclusion]This paper tests the proposed method using the field of"privacy protection"as the experimental subject,and validates the effectiveness of the interdisciplinary concept association method based on representation learning.关键词
跨学科/概念知识融合/知识表示学习/实体对齐/概念关联Key words
interdisciplinary/conceptual knowledge fusion/knowledge representation learning/entity alignment/conceptual association分类
社会科学引用本文复制引用
黄京,张光照,王忠义..基于表示学习的跨学科概念关联研究[J].现代情报,2026,46(2):172-184,13.基金项目
教育部人文社会科学研究规划基金"跨学科知识组织中学科概念跨学科关联研究"(项目编号:21YJA870003). （项目编号:21YJA870003）
