AgraphG=(V, E)isasupportof a hypergraphH=(V, S)if every hyperedge induces a connected subgraph inG. Supports are usedfor certain types of hypergraph visualizations. In this paper we considervisualizingspatialhypergraphs, where each vertex has a fixed location inthe plane. This is the case, e.g., when modeling set systems of geospatiallocations as hypergraphs. By applying established aesthetic quality cri-teria we are interested in finding supports that yield plane straight-linedrawings with minimum total edge length on the input point setV.Wefirst show, from a theoretical point of view, that the problem isNP-hardalready under rather mild conditions as well as a negative approxima-bility results. Therefore, the main focus of the paper lies on practicalheuristic algorithms as well as an exact, ILP-based approach for comput-ing short plane supports. We report results from computational exper-iments that investigate the effect of requiring planarity and acyclicityon the resulting support length. Further, we evaluate the performanceand trade-offs between solution quality and speed of several heuristicsrelative to each other and compared to optimal solutions.
%0 Book Section
%1 journals/jgaa/CastermansGMNY19
%A Castermans, Thom
%A van Garderen, Mereke
%A Meulemans, Wouter
%A Nöllenburg, Martin
%A Yuan, Xiaoru
%D 2019
%E Biedl, T.
%E Kerren, A.
%I Springer International Publishing
%J Graph Drawing and Network Visualization. GD 2018. Lecture Notes in Computer Science
%K 2019 B02 from:leonkokkoliadis sfbtrr161 visus visus:garderme
%P 53-66
%R 10.1007/978-3-030-04414-5_4
%T Short Plane Supports for Spatial Hypergraphs
%U https://link.springer.com/chapter/10.1007/978-3-030-04414-5_4#citeas
%V 11282
%X AgraphG=(V, E)isasupportof a hypergraphH=(V, S)if every hyperedge induces a connected subgraph inG. Supports are usedfor certain types of hypergraph visualizations. In this paper we considervisualizingspatialhypergraphs, where each vertex has a fixed location inthe plane. This is the case, e.g., when modeling set systems of geospatiallocations as hypergraphs. By applying established aesthetic quality cri-teria we are interested in finding supports that yield plane straight-linedrawings with minimum total edge length on the input point setV.Wefirst show, from a theoretical point of view, that the problem isNP-hardalready under rather mild conditions as well as a negative approxima-bility results. Therefore, the main focus of the paper lies on practicalheuristic algorithms as well as an exact, ILP-based approach for comput-ing short plane supports. We report results from computational exper-iments that investigate the effect of requiring planarity and acyclicityon the resulting support length. Further, we evaluate the performanceand trade-offs between solution quality and speed of several heuristicsrelative to each other and compared to optimal solutions.
@inbook{journals/jgaa/CastermansGMNY19,
abstract = {AgraphG=(V, E)isasupportof a hypergraphH=(V, S)if every hyperedge induces a connected subgraph inG. Supports are usedfor certain types of hypergraph visualizations. In this paper we considervisualizingspatialhypergraphs, where each vertex has a fixed location inthe plane. This is the case, e.g., when modeling set systems of geospatiallocations as hypergraphs. By applying established aesthetic quality cri-teria we are interested in finding supports that yield plane straight-linedrawings with minimum total edge length on the input point setV.Wefirst show, from a theoretical point of view, that the problem isNP-hardalready under rather mild conditions as well as a negative approxima-bility results. Therefore, the main focus of the paper lies on practicalheuristic algorithms as well as an exact, ILP-based approach for comput-ing short plane supports. We report results from computational exper-iments that investigate the effect of requiring planarity and acyclicityon the resulting support length. Further, we evaluate the performanceand trade-offs between solution quality and speed of several heuristicsrelative to each other and compared to optimal solutions.},
added-at = {2020-10-09T12:31:48.000+0200},
author = {Castermans, Thom and van Garderen, Mereke and Meulemans, Wouter and Nöllenburg, Martin and Yuan, Xiaoru},
biburl = {https://puma.ub.uni-stuttgart.de/bibtex/290662e67b3c6064e2bf5537fcb9d65f8/mueller},
description = {Short Plane Supports for Spatial Hypergraphs},
doi = {10.1007/978-3-030-04414-5_4},
editor = {Biedl, T. and Kerren, A.},
ee = {https://doi.org/10.7155/jgaa.00499},
interhash = {ab37db2e94a0653d0494ed37c215f967},
intrahash = {90662e67b3c6064e2bf5537fcb9d65f8},
journal = { Graph Drawing and Network Visualization. GD 2018. Lecture Notes in Computer Science},
keywords = {2019 B02 from:leonkokkoliadis sfbtrr161 visus visus:garderme},
pages = {53-66},
publisher = {Springer International Publishing},
timestamp = {2020-10-09T10:31:48.000+0200},
title = {Short Plane Supports for Spatial Hypergraphs},
url = {https://link.springer.com/chapter/10.1007/978-3-030-04414-5_4#citeas},
volume = 11282,
year = 2019
}