基于多特征融合及奖惩机制的藏医药领域实体关系联合抽取
Joint Extraction of Entity Relations in the Field of Tibetan Medicine Based on Multi-Feature Fusion and Reward-and-Punishment Mechanism
于韬 1拥措 1高兴 1尼玛扎西1
作者信息
- 1. 西藏大学 信息科学技术学院,西藏 拉萨 850000;西藏自治区藏文信息技术人工智能重点实验室,西藏 拉萨 850000;藏文信息技术教育部工程研究中心,西藏 拉萨 850000
- 折叠
摘要
实体关系联合抽取任务旨在识别命名实体的同时可抽取实体间的语义关系.该文提出了一种基于多特征融合及奖惩机制的藏医药领域实体关系联合抽取方法,针对基于序列标注的联合抽取方法中标注策略的局限性及特征单一、模型学习能力有限的问题,提出以下解决方案:①使用嵌套实体标注策略突破原有标注方法的局限;②使用类别特征静态融合、多特征动态融合方法及奖惩机制分别用于特征增强及模型优化.实验结果表明,该文方法提升了藏医药领域联合抽取模型的效果,模型最终的F1 值为 79.23%.同时,为了证明该文模型的鲁棒性及有效性,还在 SKE及 NYT领域数据上进行了相关实验,实验结果验证该模型的有效性,且优于基线方法.
Abstract
The entity relation joint extraction task refers to extracting semantic relations between entities while iden-tifying named entities.This paper proposes a joint extraction method of entity relations in the Tibetan medicine field based on multi feature fusion and reward-and-punishment mechanism.We adopt the nested entity annotation strategy to break through the limitations of existing annotation methods.The static fusion of category features,dynamic fusion of multi features,and reward-and-punishment mechanisms are applied for feature enhancement and model optimization.The experimental results show that our method is effective and superior to the baseline methods.
关键词
藏医药/实体关系/联合抽取/多特征融合/奖惩机制Key words
Tibetan medicine/entity relation/joint extraction/multi-feature fusion/reward and punishment mecha-nism引用本文复制引用
基金项目
科技创新新一代人工智能重大项目(2030)(2022ZD0116100)
西藏自治区科技厅项目()
出版年
2024