ethical-spectacle/gus-dataset-v1

---
license: mit
task_categories:
- token-classification
language:
- en
tags:
- Social Bias
pretty_name: GUS Dataset
size_categories:
- 1K<n<10K
---

This dataset contains word level (not token-level) multi-label annotations for named-entity-recognition of:

- Generalizations (B-GEN, I-GEN)
- Unfairness (B-UNFAIR, I-UNFAIR)
- Stereotypes (B-STEREO, I-STEREO)
- Neutral (O)

The labels use Beginning, Inside, and Outside format (BIO). This, in combination with the multi-label approach allow for nested and overlapping entities that span multiple words.

[💻 NER Annotation Pipeline](https://github.com/Ethical-Spectacle/bias-detection/blob/main/NER/ner_annotation_pipeline.ipynb) used to label this dataset.

---

## GUS-Net Project Details:

#### Resources:

- Please visit this [collection](https://huggingface.co/collections/ethical-spectacle/gus-net-66edfe93801ea45d7a26a10f) for the datasets and model presented in the [GUS-Net paper](https://huggingface.co/papers/2410.08388).
- GUS-Net was implemented as part of [The Fair-ly Project](https://ethical-spectacle-research.gitbook.io/fair-ly), in a [Chrome Extension](https://chromewebstore.google.com/detail/fair-ly/geoaacpcopfegimhbdemjkocekpncfcc), and [PyPI package](https://ethical-spectacle-research.gitbook.io/fair-ly/toolkit/python-package).

#### Please cite:
```
@article{powers2024gusnet,
title={{GUS-Net: Social Bias Classification in Text with Generalizations, Unfairness, and Stereotypes}},
author={Maximus Powers and Umang Mavani and Harshitha Reddy Jonala and Ansh Tiwari and Hua Wei},
journal={arXiv preprint arXiv:2410.08388},
year={2024},
url={https://arxiv.org/abs/2410.08388}
}
```

Give our research group, [Ethical Spectacle](https://huggingface.co/ethical-spectacle), a follow ;).