naist-nlp/wned
收藏Hugging Face2025-12-10 更新2026-01-03 收录
下载链接：
https://hf-mirror.com/datasets/naist-nlp/wned
下载链接
链接失效反馈官方服务：
资源简介：
---
language:
- en
configs:
- config_name: wiki
data_files:
- split: test
path: data/wikipedia-00001-of-00001.jsonl
- config_name: cweb
data_files:
- split: test
path: data/cweb-00001-of-00001.jsonl
---
# Dataset Card for Reddit-EL
The WNED (Walking Named Entity Disambiguation) is created from Wikipedia and Web texts (ClueWeb12).
As this dataset is provided with only Wikipedia page titles, we acquired the Wikipedia page IDs via [ZELDA](https://github.com/flairNLP/zelda).
## Dataset Description
- **Homepage:** [https://github.com/U-Alberta/wned](https://github.com/U-Alberta/wned)
- **Public:** True
- **Source:** Wikipedia, ClueWeb
- **Paper:** [Robust Entity Linking via Random Walks](https://dl.acm.org/doi/10.1145/2661829.2661887)
- **Target KB** Wikipedia
### Citation Information
```
@inproceedings{DBLP:conf/cikm/GuoB14,
author = {Zhaochen Guo and Denilson Barbosa},
title = {Robust Entity Linking via Random Walks},
booktitle = {Proceedings of the 23rd {ACM} International Conference on Conference
on Information and Knowledge Management, {CIKM} 2014, Shanghai, China,
November 3-7, 2014},
pages = {499--508},
year = {2014},
url = {http://doi.acm.org/10.1145/2661829.2661887},
doi = {10.1145/2661829.2661887}
}
```
提供机构：
naist-nlp
