Corpus ID: 14199015

Semantics derived automatically from language corpora necessarily contain human biases

@article{Caliskan2016SemanticsDA,
  title={Semantics derived automatically from language corpora necessarily contain human biases},
  author={Aylin Caliskan and J. Bryson and A. Narayanan},
  journal={ArXiv},
  year={2016},
  volume={abs/1608.07187}
}
Artificial intelligence and machine learning are in a period of astounding growth. [...] Key Result These regularities are captured by machine learning along with the rest of semantics. In addition to our empirical findings concerning language, we also contribute new methods for evaluating bias in text, the Word Embedding Association Test (WEAT) and the Word Embedding Factual Association Test (WEFAT).Expand
89 Citations
Using word embeddings to generate data-driven human agent decision-making from natural language
  • 3
  • PDF
Think Piece: On the Cruelty of Really Writing a History of Machine Learning
  • 1
Bias in machine learning - what is it good for?
  • 1
  • PDF
A General Methodology to Quantify Biases in Natural Language Data
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 59 REFERENCES
Extracting Semantics from the Enron Corpus
  • 2
A Neural Probabilistic Language Model
  • 4,900
  • PDF
Word and Object
  • 6,366
Glove: Global Vectors for Word Representation
  • 17,484
  • Highly Influential
  • PDF
Artificial Intelligence's White Guy Problem
  • 110
  • PDF
Distributed Representations of Words and Phrases and their Compositionality
  • 21,590
  • PDF
Big Data's Disparate Impact
  • 1,049
  • PDF
The native language of social cognition
  • 558
  • PDF
...
1
2
3
4
5
...