20newsW100
Description
Nodes represent news articles. Each hyperedge represents a word and contains the set of articles in which that word occurs. A dataset from the UCI Categorical Machine Learning Repository. The dataset contains 16,242 articles with binary occurrence values of 100 words. Each word is regarded as a hyperedge and the news articles are vertices. Dataset construction following Yang et al. (Chaoqi Yang, Ruijie Wang, Shuochao Yao, and Tarek Abdelzaher. Hypergraph learning with line expansion. arXiv preprint arXiv:2005.04843, 2020.).
Basic statistics
- Nodes: 16242
- Hyperedges: 100
- Unique hyperedges: 100
- Max size hyperedge: 2241
Hypergraph metadata
| Property | Description |
|---|---|
| type | (STRING) Hypergraph type (e.g., Hypergraph). |
| weighted | (BOOL) Whether the hypergraph is weighted (e.g., false). |
Hyperedge size distribution
Hyperdegree distribution
Download
- Version 1.0.0 Binary (240.4 KB) JSON (179.1 KB)
Provenance
Source: https://github.com/jianhao2016/AllSet?tab=readme-ov-file
License: MIT
Reproducibility: Instructions and scripts
Citation
When this data is used in published research or for visualization purposes, please cite the following:
Copied!
@inproceedings{chien2022you,
title={You are AllSet: A Multiset Function Framework for Hypergraph Neural Networks},
author={Eli Chien and Chao Pan and Jianhao Peng and Olgica Milenkovic},
booktitle={International Conference on Learning Representations},
year={2022},
url={https://openreview.net/forum?id=hpBTIv2uy_E}
}