Quantitative Comparison of Time‐Dependent Treemaps

Abstract

Rectangular treemaps are often the method of choice to visualize large hierarchical datasets. Nowadays such datasets are available over time, hence there is a need for (a) treemaps that can handle time‐dependent data, and (b) corresponding quality criteria that cover both a treemap's visual quality and its stability over time. In recent years a wide variety of (stable) treemapping algorithms has been proposed, with various advantages and limitations. We aim to provide insights to researchers and practitioners to allow them to make an informed choice when selecting a treemapping algorithm for specific applications and data. To this end, we perform an extensive quantitative evaluation of rectangular treemaps for time‐dependent data. As part of this evaluation we propose a novel classification scheme for time‐dependent datasets. Specifically, we observe that the performance of treemapping algorithms depends on the characteristics of the datasets used. We identify four potential representative features that characterize time‐dependent hierarchical datasets and classify all datasets used in our experiments accordingly. We experimentally test the validity of this classification on more than 2000 datasets, and analyze the relative performance of 14 state‐of‐the‐art rectangular treemapping algorithms across varying features. Finally, we visually summarize our results with respect to both visual quality and stability to aid users in making an informed choice among treemapping algorithms. All datasets, metrics, and algorithms are openly available to facilitate reuse and further comparative studies.

SplitStreams: A Visual Metaphor for Evolving Hierarchies

“…In order to evaluate the performance of our current implementation, we ran the algorithm for generating SplitStreams (Appendix A) for a subset (1313 evolving hierarchies) of Vernier et al [21]'s benchmark datasets ranging from small datasets up to approximately 100K nodes. The results are plotted in Figure 5 (right) and the table on the left shows exact timings for selected representative datasets.…”

“…Several works have stabilized the positioning of individual data items in treemaps and optimized the layout strategy for easier comparison and tracking of items [4], [5], [6], [7], [8], [9]. Vernier et al [21] performed a quantitative comparison of 13 different treemap layouts and collected a benchmark dataset consisting of 2720 evolving hierarchies for that purpose. While juxtaposed trees provide a good understanding of hierarchical structures at a given point in time, they lack a clear representation of time-related changes.…”

“…Vernier et al [21] collected 2720 datasets showing the evolution of values over time, roughly half of which feature a hierarchy.…”

“…Being able to see a clear separation at every timestep further provides us with a better intuition for the time scale being used. All benchmark datasets from Vernier et al [21] can be investigated through the exemplary implementation of our method in the supplemental material.…”

Treemap Layout Based on Monte Carlo Tree Search

“…其他形状 [1][2][3] , 矩形较为规则、易于实现, 可以完全 利用所给的矩形绘制区域, 空间填充型是层次数 据可视化方法中最常使用矩形. 目前常用的矩形树图布局算法有 14 种 [4] . 评 估树图的方式有平均长宽比、数据一致性、数据稳 定性和可读性 [5] 本文根据树图布局原则对树图布局算法进行 分类, 发现符合逐步加点布局原则的 3 种树图布局 算法：正方化布局算法(squarified, SQR) [6] 、条状布 局 算 法 (strip, STR) [7] 和 螺 旋 线 布 局 算 法 (spiral, SPI) [5] 1.1 基于长宽比优化的树图布局算法 1992 年, Shneiderman [8] 第 1 次提出用区域嵌套 的方式展示树结构, 并设计了横切和纵切布局算 法(slice and dice, SND).…”