• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Computer speech and language >Controlling contents in data-to-document generation with human-designed topic labels
【24h】

Controlling contents in data-to-document generation with human-designed topic labels

机译：使用人为设计主题标签控制数据到文档生成中的内容

获取原文
获取原文并翻译 | 示例
       
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

We propose a data-to-document generator that can easily control the contents of output texts based on a neural language model. Conventional data-to-text model is useful when a reader seeks a global summary of data because it has only to describe an important part that has been extracted beforehand. However, since it differs from users to users what they are interested in, it is necessary to develop a method to generate various summaries according to users' requests. We develop a model to generate various summaries and to control their contents by providing the explicit targets for a reference to the model as controllable factors. In the experiments, we used five-minute or one-hour charts of 9 indicators (e.g., Nikkei 225), as time-series data, and daily summaries of Nikkei Quick News as textual data. We conducted comparative experiments using two pieces of information: human-designed topic labels indicating the contents of a sentence and automatically extracted keywords as the referential information for generation. Experiments show that both models using additional information of target document achieved higher performance in terms of BLEU and human evaluation. We found that human-designed topic labels are superior to extracted keywords in terms of controllability.
机译：我们提出了一种数据到文档生成器，可以根据神经语言模型轻松控制输出文本的内容。当读者寻求全局数据摘要时，传统的数据到文本模型非常有用，因为它只用于描述事先提取的重要部分。但是，由于它与用户不同于用户感兴趣的用户，因此有必要根据用户的请求开发一种生成各种摘要的方法。我们开发一个模型来生成各种摘要，并通过提供对模型作为可控因素的参考的显式目标来控制其内容。在实验中，我们使用了5分钟或一小时的9个指标（例如，日经225）图表，作为时间序列数据，日经Nikkei快速新闻的日常摘要作为文本数据。我们使用两条信息进行了比较实验：人类设计的主题标签，指示句子的内容并自动提取关键字作为生成的参考信息。实验表明，两种模型，使用目标文件的其他信息在BLEU和人类评估方面取得了更高的性能。我们发现，在可控性方面，人类设计的主题标签优于提取的关键字。

著录项

  • 来源
    《Computer speech and language》 |2021年第3期|101154.1-101154.10|共10页
  • 作者

    Kasumi Aoki; Akira Miyazawa; Tatsuya Ishigaki; Tatsuya Aoki; Hiroshi Noji; Keiichi Goshima; Hiroya Takamura; Yusuke Miyao; Ichiro Kobayashi;

  • 作者单位

    Ochanomizu University Japan National Institute of Advanced Industrial Science and Technology Japan;

    The Graduate University for Advanced Studies Japan National Institute of Informatics Japan National Institute of Advanced Industrial Science and Technology Japan;

    National Institute of Advanced Industrial Science and Technology Japan;

    Tokyo Institute of Technology Japan National Institute of Advanced Industrial Science and Technology Japan;

    National Institute of Advanced Industrial Science and Technology Japan;

    Waseda University Japan National Institute of Advanced Industrial Science and Technology Japan;

    Tokyo Institute of Technology Japan National Institute of Advanced Industrial Science and Technology Japan;

    The University of Tokyo Japan National Institute of Advanced Industrial Science and Technology Japan;

    Ochanomizu University Japan National Institute of Advanced Industrial Science and Technology Japan;

  • 收录信息 美国《科学引文索引》(SCI);美国《工程索引》(EI);
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

    Natural language generation; Data-to-text; Time-series data; Topic guided controllability;

    机译：自然语言生成;数据到文本;时间序列数据;主题导向可控性;

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