Datasets:
metadata
annotations_creators:
- no-annotation
language:
- fa
- ru
- zh
language_creators:
- found
license:
- odc-by
multilinguality:
- multilingual
pretty_name: HC4
size_categories:
- 1M<n<10M
source_datasets:
- extended|c4
tags: []
task_categories:
- text-retrieval
task_ids:
- document-retrieval
Dataset Card for HC4
Dataset Description
- Repository: https://github.com/hltcoe/HC4
- Paper: https://arxiv.org/abs/2201.09992
Dataset Summary
HC4 is a suite of test collections for ad hoc Cross-Language Information Retrieval (CLIR), with Common Crawl News documents in Chinese, Persian, and Russian. The documents are Web pages from Common Crawl in Chinese, Persian, and Russian.
Languages
- Chinese
- Persian
- Russian
Dataset Structure
Data Instances
Split | Documents |
---|---|
fas (Persian) |
486K |
rus (Russian) |
4.7M |
zho (Chinese) |
646K |
Data Fields
id
: unique identifier for this documentcc_file
: source file from connon crawltime
: extracted date/time from articletitle
: title extracted from articletext
: extracted article bodyurl
: source URL
Dataset Usage
Using 🤗 Datasets:
from datasets import load_dataset
dataset = load_dataset('neuclir/hc4')
dataset['fas'] # Persian documents
dataset['rus'] # Russian documents
dataset['zho'] # Chinese documents
Citation Information
@article{Lawrie2022HC4,
author = {Dawn Lawrie and James Mayfield and Douglas W. Oard and Eugene Yang},
title = {HC4: A New Suite of Test Collections for Ad Hoc CLIR},
booktitle = {{Advances in Information Retrieval. 44th European Conference on IR Research (ECIR 2022)},
year = {2022},
month = apr,
publisher = {Springer},
series = {Lecture Notes in Computer Science},
site = {Stavanger, Norway},
url = {https://arxiv.org/abs/2201.09992}
}