Graph Benchmark Datasets

The following is a list of benchmark datasets for testing graph layout algorithms. The list was collected at the Khoury Vis Lab at Northeastern University, and is maintained by the same. Our colleciton methodology targeted layout algorithms specifically—we do acknowledge the existence of other repositories that target other network-related purposes more in detail. The collection and supplemental material is also accessible at https://osf.io/j7ucv/

Click on the names of the collections to expand them and access information about their contents and a list of papers using them.

If you find our work useful for your research, consider citing our paper as well as the linked "Origin Paper" for each dataset used:



    @Misc{DiBartolomeo2024BenchmarkDatasetsGraph,
        author       = {Di~Bartolomeo, Sara and Wilson, Connor and Puerta, Eduardo and Crnovrsanin, Tarik and Frings, Alexander and Dunne, Cody},
        howpublished = {Under submission to JoVI (\url{https://www.journalovi.org/})},
        note         = {Supplemental material at \url{osf.io/j7ucv/}.},
        title        = {Benchmark datasets for graph layout algorithms},
        year         = {2024},
        url          = {https://visdunneright.github.io/gd_benchmark_sets/},
    }

We also have a poster for an earlier version of this work:



    @Misc{DiBartolomeo2023CollectionBenchmarkDatasets,
        author       = {Di~Bartolomeo, Sara and Puerta, Eduardo and Wilson, Connor and Crnovrsanin, Tarik and Dunne, Cody},
        howpublished = {Poster at Graph Drawing and Network Visualization '23},
        note         = {GD 2023 Best Poster Honorable Mention! Preprint at \url{https://osf.io/yftju/}. Supplemental material at \url{osf.io/j7ucv/}.},
        title        = {A collection of benchmark datasets for evaluating graph layout algorithms},
        year         = {2023},
        series       = {GD Posters},
        url          = {https://visdunneright.github.io/gd_benchmark_sets/}
      }

Data Collection Process:
The initial datasets were collected as a result of the analysis and exploration of what datasets were used in 196 graph-drawing-related papers, mostly sourced from the Graph Drawing conference's proceedings, plus searches done on IEEE Xplore and the Wiley Digital Library. A detailed description of the collection process can be found at https://www.journalovi.org/2024-dibartolomeo-benchmark/

Contributing:

Please open an issue here: https://github.com/VisDunneRight/gd_benchmark_sets using this issue template: https://github.com/VisDunneRight/gd_benchmark_sets/issues/new?template=dataset-inclusion.md.

Alternatively, reach out to dibartolomeo.sara@gmail.com

Graph Layout Benchmark Datasets