---
license: mit
pretty_name: Twitch Gamers Social Network
configs:
  - config_name: edges
    data_files:
      - split: train
        path: data/edges/*
  - config_name: features
    data_files:
      - split: train
        path: data/features/*
---

# Twitch Gamers Social Network

A social network of Twitch users which was collected from the public API in Spring 2018. Nodes are Twitch users and edges are mutual follower relationships between them. The graph forms a single strongly connected component without missing attributes. The machine learning tasks related to the graph are count data regression and node classification. There are 6 specific tasks:

- Explicit content streamer identification.
- Broadcaster language prediction.
- User lifetime estimation.
- Churn prediction.
- Affiliate status identification.
- View count estimation.

https://snap.stanford.edu/data/twitch_gamers.html


| Dataset statistics |           |
|--------------------|-----------|
| Directed           | No.       |
| Node features      | No.       |
| Edge features      | No.       |
| Node labels        | Yes.      |
| Temporal           | No.       |
| Nodes              | 168,114   |
| Edges              | 6,797,557 |
| Density            | 0.0005    |
| Transitvity        | 0.0184    |

```bibtex
@misc{rozemberczki2021twitch,
      title={Twitch Gamers: a Dataset for Evaluating Proximity Preserving and Structural Role-based Node Embeddings}, 
      author={Benedek Rozemberczki and Rik Sarkar},
      year={2021},
      eprint={2101.03091},
      archivePrefix={arXiv},
      primaryClass={cs.SI}
 }
```

## Dataset Description

- **Homepage:** https://github.com/benedekrozemberczki/datasets
- **Source**: https://snap.stanford.edu/data/twitch_gamers.html
- **Author:** B. Rozemberczki and R. Sarkar
- **Created:** 2021