Quantitative data visualization with interactive KDE surfaces
Abstract
Kernel density estimation (KDE) is an established statistical concept for assessing the distributional characteristics of data that also has proven its usefulness for data visualization. In this work,we present several enhancements to such a KDE-based visualization that aim (a) at an improved specificity of the visualization with respect to the communication of quantitative information about the data and its distribution and (b) at an improved integration of such a KDE-based visualization in an interactive visualization setting, where, for example, linking and brushing is easily possible both from and to such a visualization. With our enhancements to KDE-based visualization, we can extend the utilization of this great statistical concept in the context of interactive visualization.
M. Florek and H. Hauser, "Quantitative data visualization with interactive KDE surfaces," in Proceedings of the Spring Conference on Computer Graphics (SCCG 2010), 2010, p. –.
[BibTeX]
Kernel density estimation (KDE) is an established statistical concept for assessing the distributional characteristics of data that also has proven its usefulness for data visualization. In this work,we present several enhancements to such a KDE-based visualization that aim (a) at an improved specificity of the visualization with respect to the communication of quantitative information about the data and its distribution and (b) at an improved integration of such a KDE-based visualization in an interactive visualization setting, where, for example, linking and brushing is easily possible both from and to such a visualization. With our enhancements to KDE-based visualization, we can extend the utilization of this great statistical concept in the context of interactive visualization.
@INPROCEEDINGS {florek10kde,
author = "Martin Florek and Helwig Hauser",
title = "Quantitative data visualization with interactive KDE surfaces",
booktitle = "Proceedings of the Spring Conference on Computer Graphics (SCCG 2010)",
year = "2010",
pages = "--",
month = "May",
abstract = "Kernel density estimation (KDE) is an established statistical concept for assessing the distributional characteristics of data that also has proven its usefulness for data visualization. In this work,we present several enhancements to such a KDE-based visualization that aim (a) at an improved specificity of the visualization with respect to the communication of quantitative information about the data and its distribution and (b) at an improved integration of such a KDE-based visualization in an interactive visualization setting, where, for example, linking and brushing is easily possible both from and to such a visualization. With our enhancements to KDE-based visualization, we can extend the utilization of this great statistical concept in the context of interactive visualization.",
images = "images/florek10kde.jpg",
thumbnails = "images/florek10kde_thumb.jpg",
location = "Budmerice, Slovakia",
url = "//dx.doi.org/10.1145/1925059.1925068"
}