README.md

metadata

license: mit
language:
  - en
library_name: transformers

PoliticalBiasBERT

BERT finetuned on many examples of politically biased texts

Paper and repository coming soon.

Usage

from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch

text = "your text here"

tokenizer = AutoTokenizer.from_pretrained("bert-base-cased")

model = AutoModelForSequenceClassification.from_pretrained("bucketresearch/politicalBiasBERT")


inputs = tokenizer(text, return_tensors="pt")
labels = torch.tensor([0])
outputs = model(**inputs, labels=labels)
loss, logits = outputs[:2]

# [0] -> left 
# [1] -> center
# [2] -> right
print(logits.softmax(dim=-1)[0].tolist()) 

References

@inproceedings{baly2020we,
  author      = {Baly, Ramy and Da San Martino, Giovanni and Glass, James and Nakov, Preslav},
  title       = {We Can Detect Your Bias: Predicting the Political Ideology of News Articles},
  booktitle   = {Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)},
  series      = {EMNLP~'20},
  NOmonth     = {November},
  year        = {2020}
  pages       = {4982--4991},
  NOpublisher = {Association for Computational Linguistics}
}

@article{bucket_bias2023,
  organization={Bucket Research}
  title={Political Bias Classification using finetuned BERT model}
  year={2023}

}