Force-Directed Graph Layouts by Edge Sampling

Recent work shows that sampling algorithms can be an effective tool for graph visualization. This paper extends prior work by applying edge sampling algorithms to speed up the spring force calculation in force-directed graph layout algorithms. An experiment on 72 graphs finds that some sampling algo...

Full description

Saved in:
Bibliographic Details
Published in:2019 IEEE 9th Symposium on Large Data Analysis and Visualization (LDAV) pp. 1 - 5
Main Author: Gove, Robert
Format: Conference Proceeding
Language:English
Published: IEEE 01.10.2019
Subjects:
Approximation algorithms
Force
Human-centered computing-Visualization-Visualizationtechniques-Graph drawings
Layout
Microsoft Windows
Readability metrics
Runtime
Sketching and sampling
Springs
sublinear and near linear time algorithms-Sketching and sampling
Theory of computtion-Design and analysis of algorithms-Streaming
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Recent work shows that sampling algorithms can be an effective tool for graph visualization. This paper extends prior work by applying edge sampling algorithms to speed up the spring force calculation in force-directed graph layout algorithms. An experiment on 72 graphs finds that some sampling algorithms achieve comparable quality as no sampling. This result is confirmed with visualizations of the graph layout results. However, runtime improvements are small, especially for graphs with 10,000 vertices or fewer, indicating that the runtime savings might not be worth the risk to layout quality. Therefore, this paper suggests that accurate spring forces may be more important to force-directed graph layout algorithms than accurate electric forces. A copy of this paper plus the code and data to reproduce the results are available at https://osf.io/4ja29/.
DOI:10.1109/LDAV48142.2019.8944364