Evaluating Representation Learning and Graph Layout Methods for Visualization

Graphs and other structured data have come to the forefront in machine learning over the past few years due to the efficacy of novel representation learning methods boosting the prediction performance in various tasks. Representation learning methods embed the nodes in a low-dimensional real-valued...

Ausführliche Beschreibung

Bibliographische Detailangaben
Veröffentlicht in:IEEE computer graphics and applications. - 1991. - 42(2022), 3 vom: 01. Mai, Seite 19-28
1. Verfasser: Heiter, Edith (VerfasserIn)
Weitere Verfasser: Kang, Bo, De Bie, Tijl, Lijffijt, Jefrey, Potel, Mike
Format: Online-Aufsatz
Sprache:English
Veröffentlicht: 2022
Zugriff auf das übergeordnete Werk:IEEE computer graphics and applications
Schlagworte:Journal Article Research Support, Non-U.S. Gov't
LEADER 01000naa a22002652 4500
001 NLM341931004
003 DE-627
005 20231226012913.0
007 cr uuu---uuuuu
008 231226s2022 xx |||||o 00| ||eng c
024 7 |a 10.1109/MCG.2022.3160104  |2 doi 
028 5 2 |a pubmed24n1139.xml 
035 |a (DE-627)NLM341931004 
035 |a (NLM)35671278 
040 |a DE-627  |b ger  |c DE-627  |e rakwb 
041 |a eng 
100 1 |a Heiter, Edith  |e verfasserin  |4 aut 
245 1 0 |a Evaluating Representation Learning and Graph Layout Methods for Visualization 
264 1 |c 2022 
336 |a Text  |b txt  |2 rdacontent 
337 |a ƒaComputermedien  |b c  |2 rdamedia 
338 |a ƒa Online-Ressource  |b cr  |2 rdacarrier 
500 |a Date Completed 09.06.2022 
500 |a Date Revised 09.07.2022 
500 |a published: Print 
500 |a Citation Status MEDLINE 
520 |a Graphs and other structured data have come to the forefront in machine learning over the past few years due to the efficacy of novel representation learning methods boosting the prediction performance in various tasks. Representation learning methods embed the nodes in a low-dimensional real-valued space, enabling the application of traditional machine learning methods on graphs. These representations have been widely premised to be also suited for graph visualization. However, no benchmarks or encompassing studies on this topic exist. We present an empirical study comparing several state-of-the-art representation learning methods with two recent graph layout algorithms, using readability and distance-based measures as well as the link prediction performance. Generally, no method consistently outperformed the others across quality measures. The graph layout methods provided qualitatively superior layouts when compared to representation learning methods. Embedding graphs in a higher dimensional space and applying t-distributed stochastic neighbor embedding for visualization improved the preservation of local neighborhoods, albeit at substantially higher computational cost 
650 4 |a Journal Article 
650 4 |a Research Support, Non-U.S. Gov't 
700 1 |a Kang, Bo  |e verfasserin  |4 aut 
700 1 |a De Bie, Tijl  |e verfasserin  |4 aut 
700 1 |a Lijffijt, Jefrey  |e verfasserin  |4 aut 
700 1 |a Potel, Mike  |e verfasserin  |4 aut 
773 0 8 |i Enthalten in  |t IEEE computer graphics and applications  |d 1991  |g 42(2022), 3 vom: 01. Mai, Seite 19-28  |w (DE-627)NLM098172794  |x 1558-1756  |7 nnns 
773 1 8 |g volume:42  |g year:2022  |g number:3  |g day:01  |g month:05  |g pages:19-28 
856 4 0 |u http://dx.doi.org/10.1109/MCG.2022.3160104  |3 Volltext 
912 |a GBV_USEFLAG_A 
912 |a SYSFLAG_A 
912 |a GBV_NLM 
912 |a GBV_ILN_350 
951 |a AR 
952 |d 42  |j 2022  |e 3  |b 01  |c 05  |h 19-28