tarudesu/ViHealthQA

---
task_categories:
- question-answering
language:
- vi
tags:
- medical
pretty_name: Vietnamese Healthcare Question Answering Dataset
size_categories:
- 10K<n<100K
---
## Disclaimer:
The dataset may contain personal information crawled along with the contents of various sources. Please make a filter in pre-processing data before starting your research training.

# SPBERTQA: A Two-Stage Question Answering System Based on Sentence Transformers for Medical Texts
This is the official repository for the ViHealthQA dataset from the paper [SPBERTQA: A Two-Stage Question Answering System Based on Sentence Transformers for Medical Texts](https://arxiv.org/pdf/2206.09600.pdf), which was accepted at the [KSEM-2022](https://ksem22.smart-conf.net/index.html).

# Citation Information
The provided dataset is only used for research purposes!

```
@InProceedings{nguyen2022viheathqa,
author="Nguyen, Nhung Thi-Hong
and Ha, Phuong Phan-Dieu
and Nguyen, Luan Thanh
and Van Nguyen, Kiet
and Nguyen, Ngan Luu-Thuy",
title="SPBERTQA: A Two-Stage Question Answering System Based on Sentence Transformers for Medical Texts",
booktitle="Knowledge Science, Engineering and Management",
year="2022",
publisher="Springer International Publishing",
address="Cham",
pages="371--382",
isbn="978-3-031-10986-7"
}

```

# Abstract

Question answering (QA) systems have gained explosive attention in recent years. However, QA tasks in Vietnamese do not have many datasets. Significantly, there is mostly no dataset in the medical domain. Therefore, we built a Vietnamese Healthcare Question Answering dataset (ViHealthQA), including 10,015 question-answer passage pairs for this task, in which questions from health-interested users were asked on prestigious health websites and answers from highly qualified experts. This paper proposes a two-stage QA system based on Sentence-BERT (SBERT) using multiple negatives ranking (MNR) loss combined with BM25. Then, we conduct diverse experiments with many bag-of-words models to assess our system’s performance. With the obtained results, this system achieves better performance than traditional methods.

# Dataset
The ViHealthQA dataset is consist of 10,015 question-answer passage pairs. Note that questions are from health-interested users asked on prestigious health websites and answers are from highly qualified experts.

The dataset is divided into three parts as below:
1. Train set: 7.01K question-answer pairs
2. Valid set: 2.01 question-answer pairs
3. Test set: 993 question-answer pairs

# Contact
Please feel free to contact us by email [e-mail] if you have any further information!

原始信息汇总

数据集概述

基本信息

• 任务类别: 问答系统
• 语言: 越南语
• 标签: 医疗
• 数据集名称: Vietnamese Healthcare Question Answering Dataset
• 数据集大小: 10K<n<100K

数据集描述

• 数据集内容: 包含10,015个问答对，问题来自关注健康的用户在知名健康网站上的提问，答案来自高资质专家。
• 数据集划分:
  1. 训练集: 7,010个问答对
  2. 验证集: 2,010个问答对
  3. 测试集: 993个问答对

引用信息

@InProceedings{nguyen2022viheathqa, author="Nguyen, Nhung Thi-Hong and Ha, Phuong Phan-Dieu and Nguyen, Luan Thanh and Van Nguyen, Kiet and Nguyen, Ngan Luu-Thuy", title="SPBERTQA: A Two-Stage Question Answering System Based on Sentence Transformers for Medical Texts", booktitle="Knowledge Science, Engineering and Management", year="2022", publisher="Springer International Publishing", address="Cham", pages="371--382", isbn="978-3-031-10986-7" }