naist-nlp/zeshel

---
language:
- en

configs:
- config_name: dataset
data_files:
- split: train
path: data/train-00001-of-00001.jsonl
- split: validation
path: data/validation-00001-of-00001.jsonl
- split: test
path: data/test-00001-of-00001.jsonl
- split: heldout_train_seen
path: data/heldout_train_seen-00001-of-00001.jsonl
- split: heldout_train_unseen
path: data/heldout_train_seen-00001-of-00001.jsonl
default: true

- config_name: dictionary
data_files:
- split: kb
path: dictionary/dictionary-00001-of-00001.jsonl

size_categories:
- 100K<n<1M

license: cc-by-sa-3.0
---

# Dataset Card for ZESHEL
ZeshEL is a dataset for Zero-shot Entity Linking.

## Dataset Description
- **Repository:** [https://github.com/lajanugen/zeshel](https://github.com/lajanugen/zeshel)
- **Public:** True
- **Source:** Wikia
- **Paper:** [Zero-Shot Entity Linking by Reading Entity Descriptions](https://aclanthology.org/P19-1335/)
- **Number of Entities** 492321

### Citation Information
```
@inproceedings{logeswaran-etal-2019-zero,
title = "Zero-Shot Entity Linking by Reading Entity Descriptions",
author = {Logeswaran, Lajanugen and Chang, Ming-Wei and
Lee, Kenton and Toutanova, Kristina and
Devlin, Jacob and Lee, Honglak},
booktitle = "Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics",
year = "2019"
}
```