Datasets:
Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,44 @@
|
|
1 |
-
---
|
2 |
-
license: apache-2.0
|
3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
task_categories:
|
4 |
+
- text-classification
|
5 |
+
language:
|
6 |
+
- ko
|
7 |
+
---
|
8 |
+
|
9 |
+
# KoMultiText: Korean Multi-task Dataset for Classifying Biased Speech
|
10 |
+
|
11 |
+
## Dataset Summary
|
12 |
+
|
13 |
+
**KoMultiText** is a comprehensive Korean multi-task text dataset designed for classifying biased and harmful speech in online platforms. The dataset focuses on tasks such as **Preference Detection**, **Profanity Identification**, and **Bias Classification** across multiple domains, enabling state-of-the-art language models to perform multi-task learning for socially responsible AI applications.
|
14 |
+
|
15 |
+
### Key Features
|
16 |
+
- **Large-Scale Dataset**: Contains 150,000 comments, including labeled and unlabeled data.
|
17 |
+
- **Multi-task Annotations**: Covers Preference, Profanity, and nine distinct types of Bias.
|
18 |
+
- **Human-Labeled**: All labeled data is annotated by **five human experts** to ensure high-quality and unbiased annotations.
|
19 |
+
- **Real-world Relevance**: Collected from "Real-time Best Gallery" of [DC Inside](https://www.dcinside.com/), a popular online community in South Korea.
|
20 |
+
|
21 |
+
---
|
22 |
+
|
23 |
+
## Dataset Creation
|
24 |
+
|
25 |
+
### Source Data
|
26 |
+
- **Origin**: Comments collected from "Real-time Best Gallery" on [DC Inside](https://www.dcinside.com/).
|
27 |
+
- **Annotation Process**:
|
28 |
+
- **Human Annotation**: Five human annotators independently labeled all comments in the dataset to ensure accuracy and minimize bias.
|
29 |
+
- **Labeling Process**: Annotators followed strict guidelines to classify comments into Preference, Profanity, and nine types of Bias. Discrepancies were resolved through majority voting and discussion.
|
30 |
+
- **Dataset Composition**:
|
31 |
+
- **Labeled Data**: 40,361 comments (train/test split).
|
32 |
+
- **Unlabeled Data**: 110,000 comments for potential pretraining or unsupervised learning.
|
33 |
+
|
34 |
+
## Citation
|
35 |
+
|
36 |
+
<pre>
|
37 |
+
@misc{choi2023largescale,
|
38 |
+
title={Large-Scale Korean Text Dataset for Classifying Biased Speech in Real-World Online Services},
|
39 |
+
author={Dasol Choi and Jooyoung Song and Eunsun Lee and Jinwoo Seo and Heejune Park and Dongbin Na},
|
40 |
+
year={2023},
|
41 |
+
eprint={2310.04313},
|
42 |
+
archivePrefix={arXiv},
|
43 |
+
primaryClass={cs.CL}
|
44 |
+
}
|