Open positions
Open research positions in SNAP group are available at undergraduate, graduate and postdoctoral levels.

Twitch Ego Nets

Dataset information

The ego-nets of Twitch users who participated in the partnership program in April 2018. Nodes are users and links are friendships. The binary classification task is to predict using the ego-net whether the ego user plays a single or multple games. Players who play a single game usually have a more dense ego-net.

Properties
Number of graphs: 127,094
Directed: No.
Node features: No.
Edge features: No.
Graph labels: Yes. Binary-labeled.
Temporal: No.
StatsMinMax
Nodes 1452
Density 0.0380.967
Diameter 12

Possible tasks
Graph classification

Paper: https://arxiv.org/abs/2003.04819
Github Page: https://github.com/benedekrozemberczki/karateclub

Source (citation)

  • B. Rozemberczki, O. Kiss, R. Sarkar: An API Oriented Open-source Python Framework for Unsupervised Learning on Graphs 2019.
  •   >@misc{karateclub2020,
        title={An API Oriented Open-source Python Framework for Unsupervised Learning on Graphs},
        author={Benedek Rozemberczki and Oliver Kiss and Rik Sarkar},
        year={2020},
        eprint={2003.04819},
        archivePrefix={arXiv},
        primaryClass={cs.LG}
    }
    
      

    Files

    File Description
    twitch_egos.zipTwitch Ego Nets dataset