首页> 外文会议>IEEE International Conference on Bioinformatics and Biomedicine >Drug drug interaction extraction from literature using a skeleton long short term memory neural network
【24h】

Drug drug interaction extraction from literature using a skeleton long short term memory neural network

机译:使用骨架长期短期记忆神经网络从文献中提取药物相互作用

获取原文

摘要

Drug Drug Interactions (DDIs) can cause harmful effect. Two shared tasks, DDIExtraction 2011 and DDIExtraction 2013, have been held to promote the implementation and comparative assessment of natural language processing techniques in the field of the pharmacovigilance domain. However, few model can meanwhile achieve state-of-the-art performance on both tasks. A major reason is the lack of representation of DDI instance structure in common. Therefore, in this paper, we propose a novel method to make full use of the DDI structure based on deep learning, in which we grasp the skeleton structure of DDI instances by a skeleton long short term memory (skeleton-LSTM) network. The experimental results show that our method can achieve an F-score of 0.677 on DDIExtraction 2011 and an F-score of 0.714 on DDIExtraction 2013, both of which are state-of-the-art.
机译:药物药物相互作用(DDI)可能导致有害作用。举行了两个共享任务,DDIExtraction 2011和DDIExtraction 2013,以促进药物警戒领域中自然语言处理技术的实施和比较评估。但是,很少有模型可以同时在两个任务上实现最先进的性能。一个主要原因是缺乏通用的DDI实例结构表示。因此,在本文中,我们提出了一种基于深度学习的,充分利用DDI结构的新方法,即通过骨架长期短期记忆(skeleton-LSTM)网络掌握DDI实例的骨架结构。实验结果表明,我们的方法可以在DDIExtraction 2011上获得0.677的F评分,在DDIExtraction 2013上获得0.714的F评分,这两个都是最先进的。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号