首页|化学物质诱导疾病关系抽取:基于证据聚焦的图推理方法

化学物质诱导疾病关系抽取:基于证据聚焦的图推理方法

扫码查看
原文链接
  • NETL
  • NSTL
  • 万方数据
针对现有方法在挖掘化学物质与疾病之间的相互作用关系时存在过多地关注全局信息而忽略少量的证据线索和局部提及交互的问题,提出了一种基于证据聚焦的提及水平文档级关系抽取方法(Evidence Focused Mention U-shaped Network,EF-MUnet).该方法首先基于上下文感知策略建模提及特征,并利用二维卷积捕获邻近提及之间的局部交互;其次为避免无关上下文的干扰,提出两种证据聚焦策略ATT-EF和RL-EF,前者将相似度作为证据线索的衡量指标,后者基于强化学习利用延迟反馈无监督地学习最优证据提取策略;最后使用U-net网络捕获实体水平的全局特征,充分挖掘语义关系.实验结果表明,与已有方法相比,EF-MUnet在生物医学数据集CDR上的F1评价指标提升了 9.7％,并且对于句间关系的抽取更具有优势.此外,在抽取药物突变相互作用的数据集DMI上,EF-MUnet也取得了最高98.6％的准确率,证明了它是一种有效的生物医学关系抽取方法并具有较好的泛化能力.
Chemical-induced Disease Relation Extraction:Graph Reasoning Method Based on Evidence Focusing
To address the problem of existing methods focusing too much on global information while neglecting a small amount of evidence clues and local mention interactions when mining the interaction between chemicals and diseases,a mention level docu-ment-level relation extraction method based on evidence focusing(EF-MUnet)is proposed.This method first models mention fea-tures based on context aware strategies and captures local interactions between adjacent mentions using two-dimensional convolu-tion network.Secondly,to avoid irrelevant context interference,two evidence focusing strategies ATT-EF and RL-EF are pro-posed.The former uses similarity as a measure of evidence clues,while the latter uses reinforcement learning to unsupervised learn the optimal evidence extraction policy with the help of delayed feedback.Finally,U-net networks are used to capture global features at the entity level and fully explore semantic relationships.Experimental results show that compared with existing me-thods,EF-MUnet's F1 score improves by 9.7％on the biomedical dataset CDR,and it has more advantages in extracting inter-sentence relations.In addition,EF-MUnet achieves the highest accuracy of 98.6％on the dataset DMI for extracting interactions between drug and mutation,proving that it is an effective biomedical relation extraction method with good generalization ability.

Relation extractionEvidence focusingReinforcement learningSelf-attention mechanismBiomedicine

周雪阳、傅启明、陈建平、陆悠、王蕴哲

展开 >

苏州科技大学电子与信息工程学院 江苏苏州 215009

苏州科技大学江苏省建筑智慧节能重点实验室 江苏苏州 215009

苏州科技大学建筑与城市规划学院 江苏苏州 215009

关系抽取 证据聚焦 强化学习 自注意力机制 生物医学

国家重点研发计划国家自然科学基金国家自然科学基金江苏省高等学校自然科学研究项目江苏省重点研发计划江苏省研究生教育教学改革项目江苏省研究生科研与实践创新计划项目

2020YFC2006602621022786207232421KJA520005BE2020026KYCX23_3321

2024

10.11896/jsjkx.230800111

计算机科学
重庆西南信息有限公司（原科技部西南信息中心）

计算机科学

CSTPCD北大核心
影响因子：0.944
ISSN：1002-137X
年,卷(期)：2024.51(10)