Interactive bivariate mode tree

Date issued

2011

Journal Title

Journal ISSN

Volume Title

Publisher

Václav Skala - UNION Agency

Abstract

Kernel density estimation (KDE) is widely used statistical method to study distribution of the data. The problem with this method is choosing the right bandwidth. In our work we focus on semiautomatic bandwidth selection with the use of visual paradigm of the Mode Tree. We not only enhanced the bivariate mode tree visualization, which was only sketched by the authors, but we also developed a hybrid CPU/GPU implementation to improve the speed of the mode tree construction. Keywords: Mode Tree, Data Visualization, Kernel Density Estimation, GPU.

Description

Subject(s)

vizualizace dat, jádrový odhad hustoty

Citation

WSCG ’2011: Poster Papers Proceedings: The 19th International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision: University of West Bohemia, Plzen, Czech Republic, January 31 - February 3, 2011, p. 41-46.