Scholarlink

Medical relation extraction is a key task in natural language processing, focusing on identifying relationships between entities such as diseases, medications, and symptoms in clinical texts, thereby supporting decision-making in healthcare. However, the inherent complexity of medical language and the limited availability of annotated data present significant challenges. In this study, the GPLinker model is enhanced through integration of the ERNIE 3.0-x-Base-zh pre-trained model, application of conditional layer normalization (CLN), and introduction of adversarial learning. The enhanced model achieves F1 scores of 78.3 on DuIE2.0, 54.1 on CMeIE-V2, 63.3 on DiaKG, demonstrating robust performance in medical relation extraction.

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

GPMC: Leveraging Conditional Layer Normalization and Adversarial Learning for Enhanced Medical Relation Extraction

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Weikang Ding,
  • Jijun Tong,
  • Qingli Zhou,
  • Meizhen Tong,
  • Zhihuan Zhang

摘要

Medical relation extraction is a key task in natural language processing, focusing on identifying relationships between entities such as diseases, medications, and symptoms in clinical texts, thereby supporting decision-making in healthcare. However, the inherent complexity of medical language and the limited availability of annotated data present significant challenges. In this study, the GPLinker model is enhanced through integration of the ERNIE 3.0-x-Base-zh pre-trained model, application of conditional layer normalization (CLN), and introduction of adversarial learning. The enhanced model achieves F1 scores of 78.3 on DuIE2.0, 54.1 on CMeIE-V2, 63.3 on DiaKG, demonstrating robust performance in medical relation extraction.