https://github.com/ethical-spectacle/bias-detection
Named entity recognition and binary classification of text sequences
https://github.com/ethical-spectacle/bias-detection
Last synced: about 1 year ago
JSON representation
Named entity recognition and binary classification of text sequences
- Host: GitHub
- URL: https://github.com/ethical-spectacle/bias-detection
- Owner: Ethical-Spectacle
- Created: 2024-08-14T01:26:09.000Z (almost 2 years ago)
- Default Branch: main
- Last Pushed: 2024-08-28T06:37:42.000Z (almost 2 years ago)
- Last Synced: 2024-08-28T07:45:05.960Z (almost 2 years ago)
- Language: Jupyter Notebook
- Size: 3.18 MB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# 🔍 Bias Detection in Text
This is where you can find all the resources we used to build datasets and models for bias detection this summer.
## NER (Token-level)
💻 [Notebook](NER/ner_annotation_pipeline.ipynb) for annotating any dataset with the entities: Generalizations, Unfairness, and Stereotypes.
💻 [Notebook](NER/ner_bert_training.ipynb) for training BERT for multi-label token classification (for nested entities).
## Binary Classification (Sequence-level)
💻 [Notebook](BinaryClassification/bert_bias_binary_training.ipynb) for training BERT to classify entire text sequences with a 0-1 probability of social bias.