• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International conference on applications of natural language to information systems >Extracting Causal Relations Among Complex Events in Natural Science Literature
【24h】

Extracting Causal Relations Among Complex Events in Natural Science Literature

机译：提取自然科学文献中复杂事件之间的因果关系

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Causal relation extraction is the task of identifying and extracting the causal relations occurring in a text. We present an approach that is especially suitable for extracting relations between complex events - which are assumed to be already identified - as found in natural science literature, supporting literature-based knowledge discovery. The approach is based on supervised learning, exploiting a wide range of linguistic features. Experimental results indicate that even with a limited amount of training data, reasonable accuracy can be obtained by using a pipeline of classifiers, optimising hyper-parameters, down-weighting negative instances and applying feature selection methods.
机译：因果关系提取是识别和提取文本中发生的因果关系的任务。我们提供了一种特别适合于提取复杂事件之间的关系的方法，这些复杂事件假定已被识别，可以在自然科学文献中找到，支持基于文献的知识发现。该方法基于监督学习，利用多种语言功能。实验结果表明，即使训练数据量有限，使用分类器流水线，优化超参数，对否定实例进行加权和应用特征选择方法也可以获得合理的准确性。

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