首页> 外文会议>SIBGRAPI Conference on Graphics, Patterns and Images >Understanding Attribute Variability in Multidimensional Projections
【24h】

Understanding Attribute Variability in Multidimensional Projections

机译:了解多维投影中的属性可变性

获取原文

摘要

Multidimensional Projection techniques can help users to find patterns in multidimensional data. However, while the visualization literature is rich in techniques designed to improve the projection itself, only a handful of papers shed light into the attributes that contribute to cluster formation or the spread of projected data. In this paper, we present a web-based visualization tool that enriches multidimensional projection layout with statistical measures derived from inputted data. Given a set of regions to analyze, we used statistical measures, such as variance, to highlight relevant attributes that contribute to the points' similarities in each region. Experimental tests show that our technique can help identify important attributes and explain projected data.
机译:多维投影技术可以帮助用户找到多维数据中的模式。但是,尽管可视化文献中包含许多旨在改善投影本身的技术,但只有少数几篇论文阐明了有助于聚类形成或投影数据传播的属性。在本文中,我们提出了一种基于Web的可视化工具,该工具利用从输入数据中得出的统计量来丰富多维投影布局。给定一组要分析的区域,我们使用统计方法(例如方差)来突出显示有助于每个区域中点相似性的相关属性。实验测试表明,我们的技术可以帮助您识别重要属性并解释预计的数据。

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