首页> 外文会议>International Conference on Information and Knowledge Engineering >Probabilistic Scoring Methods to Assist Entity Resolution Systems Using Boolean Rules
【24h】

Probabilistic Scoring Methods to Assist Entity Resolution Systems Using Boolean Rules

机译:概率评分方法,协助使用布尔规则的实体解析系统

获取原文
获取外文期刊封面目录资料

摘要

This paper describes methods to provide clerical review indicators for Entity Resolution (ER) systems that use Boolean match rules. These clerical review indicators are generated by the use of a supplemental probabilistic scoring algorithm. Four scoring methods that generate a probabilistic match score are discussed. When this match score is combined with a predefined threshold, an ER system can suggest possible near matches for clerical review. This allows the user a method to more effectively address false negative resolutions made by the system.
机译:本文介绍了为使用布尔匹配规则的实体解析(ER)系统提供文书审查指标的方法。这些文书审查指标是通过使用补充概率评分算法而产生的。讨论了产生概率匹配分数的四种评分方法。当该匹配分数与预定义的阈值相结合时,ER系统可以建议在接近匹配的匹配中进行文书审查。这允许用户更有效地解决系统制作的假否定分辨率的方法。

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