首页> 外文会议>CMG 99 Proceedings >USING BOOLEANIZED DATA TO DISCOVER BETTER RELATIONSHIPS BETWEEN METRICS
【24h】

USING BOOLEANIZED DATA TO DISCOVER BETTER RELATIONSHIPS BETWEEN METRICS

机译:使用布尔化数据来发现度量之间更好的关系

获取原文

摘要

How can we find meaningful relationships between measurements? Historically, performance analysts have defined relationships and used standard statistical measures to test relationship validity. We review a technique (previously presented at CMG96) that arranges measurement data so that an analyst can query the data to observe existing relationships. The Booleanization step in the algorithm is analyzed to see how well relationships generated by the automated Booleanization methods of mean, mode, median, and clustering correlate to relationships generated by Boolean thresholds defined by an expert.
机译:我们如何找到测量之间有意义的关系?从历史上看,绩效分析师已经定义了关系并使用标准的统计量度来测试关系的有效性。我们回顾了一种安排测量数据的技术(以前在CMG96上提出),以便分析师可以查询数据以观察现有的关系。分析算法中的布尔化步骤,以查看由均值,众数,中位数和聚类的自动布尔化方法生成的关系与专家定义的布尔阈值生成的关系之间的相关性。

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