GitHub - ahmadalsharef994/Toxicity-Classification-in-texts: An approach of measuring Unintended Bias in Toxicity Classification. This approach can be used to evaluate various toxic comments and messages in various public forums and talk pages. We also demonstrate how imbalances in training data can lead to unintended bias in resulting models, and therefore potentially unfair applications. We have used a general supervised approach using State-of-art algorithms and have used various external embeddings to improve our accuracy Of the model. We demonstrate that this approach reduces the unintended bias without compromising overall model quality.

ahmadalsharef994 / Toxicity-Classification-in-texts Public

Notifications You must be signed in to change notification settings
Fork 0
Star 1

An approach of measuring Unintended Bias in Toxicity Classification. This approach can be used to evaluate various toxic comments and messages in various public forums and talk pages. We also demonstrate how imbalances in training data can lead to unintended bias in resulting models, and therefore potentially unfair applications. We have used a …

1 star 0 forks Branches Tags Activity

Star

Notifications

Branches Tags

Name		Name	Last commit message	Last commit date
Latest commit History 10 Commits
Toxic BERT plain vanilla		Toxic BERT plain vanilla
pytorch-pretrained-BERT		pytorch-pretrained-BERT
uncased_l-12_h-768_a-12/uncased_l-12_h-768_a-12		uncased_l-12_h-768_a-12/uncased_l-12_h-768_a-12
Prerequisites.txt		Prerequisites.txt
Toxicity_Classification.ipynb		Toxicity_Classification.ipynb
bert_config.json		bert_config.json
bert_files.zip		bert_files.zip
jigsaw-starter-blend.ipynb		jigsaw-starter-blend.ipynb
sample_submission.csv		sample_submission.csv
test.csv		test.csv
train.csv		train.csv