• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>IEEE International Conference on Tools with Artificial Intelligence >NEREA: Named Entity Recognition and Disambiguation Exploiting Local Document Repositories
【24h】

NEREA: Named Entity Recognition and Disambiguation Exploiting Local Document Repositories

机译：NEREA：利用本地文档存储库的命名实体识别和歧义消除

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

摘要

In this work, we describe the design, development, and deployment of NEREA (Named Entity Recognizer for spEcific Areas), an automatic Named Entity Recognizer and Disambiguation system, developed in collaboration with professional documentalists. The aim of NEREA is to keep accurate and current information about the entities mentioned in a local repository, and then support building appropriate infoboxes, setting out the main data of these entities. It achieves a high performance thanks to the use of classification resources belonging to the local database. With this aim, the system performs tasks of named entity recognition and disambiguation by using three types of knowledge bases: local classification resources, global databases like DBpedia, and its own catalog created by NEREA. The proposed method has been validated with two different datasets and its operation has been tested in English and Spanish. The working methodology is being applied in a real environment of a media with promising results.
机译：在这项工作中，我们描述了NEREA（特殊区域的命名实体识别器）的设计，开发和部署，NEREA是与专业文献工作者合作开发的自动命名实体识别和消歧系统。 NEREA的目的是保持有关本地存储库中提到的实体的准确和最新信息，然后支持构建适当的信息框，列出这些实体的主要数据。由于使用了属于本地数据库的分类资源，因此可以实现较高的性能。为此，系统通过使用三种类型的知识库执行命名实体识别和歧义消除的任务：本地分类资源，全局数据库（如DBpedia）以及由NEREA创建的自己的目录。所提出的方法已通过两个不同的数据集进行了验证，并且其操作已通过英语和西班牙语进行了测试。该工作方法正在媒体的真实环境中应用，并取得了可喜的结果。

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