Hyukkyu/beir-trec-covid

---
license: apache-2.0
task_categories:
- information-retrieval
- text-retrieval
tags:
- beir
- trec-covid
- information-retrieval
- retrieval
- search
dataset_info:
- config_name: corpus
features:
- name: _id
dtype: string
- name: title
dtype: string
- name: text
dtype: string
- name: metadata
struct:
- name: pubmed_id
dtype: string
- name: url
dtype: string
splits:
- name: train
num_bytes: 210286181
num_examples: 171332
download_size: 115532524
dataset_size: 210286181
- config_name: queries
features:
- name: _id
dtype: string
- name: text
dtype: string
- name: metadata
struct:
- name: narrative
dtype: string
- name: query
dtype: string
splits:
- name: train
num_bytes: 13952
num_examples: 50
download_size: 11365
dataset_size: 13952
configs:
- config_name: corpus
data_files:
- split: train
path: corpus/train-*
- config_name: queries
data_files:
- split: train
path: queries/train-*
---

# BEIR TREC-COVID Dataset (Migrated)

This is a migrated version of BeIR/trec-covid that is compatible with datasets library 4.0.0+.

## Dataset Description

This dataset contains the trec-covid dataset from the BEIR benchmark, converted from the old script-based format to Parquet format.

## Dataset Structure

### Queries

- **Split 'queries'**: 50 examples
- Features: ['_id', 'text', 'metadata']

- **Total examples**: 50

### Corpus

- **Split 'corpus'**: 171,332 examples
- Features: ['_id', 'title', 'text', 'metadata']

- **Total examples**: 171,332

## Usage

```python
from datasets import load_dataset

# Load queries (split: queries)
queries = load_dataset("Hyukkyu/beir-trec-covid", "queries", split="queries")

# Load corpus (split: corpus)
corpus = load_dataset("Hyukkyu/beir-trec-covid", "corpus", split="corpus")
```

## Available Splits

### Queries
- `queries`: 50 examples

### Corpus
- `corpus`: 171,332 examples

## Original Dataset

This dataset is migrated from: BeIR/trec-covid

## Citation

If you use this dataset, please cite the original BEIR paper:

```bibtex
@article{thakur2021beir,
title={BEIR: A Heterogeneous Benchmark for Zero-shot Evaluation of Information Retrieval Models},
author={Thakur, Nandan and Reimers, Nils and Ruckle, Andreas and Srivastava, Abhishek and Gurevych, Iryna},
journal={arXiv preprint arXiv:2104.08663},
year={2021}
}
```