Research

Semantics derived automatically from language corpora contain human-like biases


Reference:

Caliskan, A., Bryson, J. J. and Narayanan, A., 2017. Semantics derived automatically from language corpora contain human-like biases. Science, 356 (6334), pp. 183-186.

Related documents:

[img]
Preview
PDF (CaliskanEtAl-authors-full) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (215kB) | Preview

    Official URL:

    https://doi.org/10.1126/science.aal4230

    Abstract

    Machine learning is a means to derive artificial intelligence by discovering patterns in existing data. Here we show that applying machine learning to ordinary human language results in human-like semantic biases. We replicate a spectrum of known biases, as measured by the Implicit Association Test, using a widely used, purely statistical machine-learning model trained on a standard corpus of text from the Web. Our results indicate that text corpora contain recoverable and accurate imprints of our historic biases, whether morally neutral as towards insects or flowers, problematic as towards race or gender, or even simply veridical, reflecting the status quo distribution of gender with respect to careers or first names. Our methods hold promise for identifying and addressing sources of bias in culture, including technology.

    Details

    Item Type Articles
    CreatorsCaliskan, A., Bryson, J. J. and Narayanan, A.
    DOI10.1126/science.aal4230
    DepartmentsFaculty of Science > Computer Science
    Research Centres & Institutes > Institute for Policy Research
    Research CentresEPSRC Centre for Doctoral Training in Statistical Mathematics (SAMBa)
    Centre for Nanoscience and Nanotechnology
    Centre for Mathematical Biology
    Publisher Statementaal4320_corrected_manuscript.pdf: This is the author’s version of the work. It is posted here by permission of the AAAS for personal use, not for redistribution. The definitive version was published in Science on Vol. 356, Issue 6334, 14 April 2017, DOI: 10.1126/science.aal4230 ;CaliskanEtAl_authors_full.pdf: This is the author’s version of the work. It is posted here by permission of the AAAS for personal use, not for redistribution. The definitive version was published in Science on Vol. 356, Issue 6334, 14 April 2017, DOI: 10.1126/science.aal4230
    RefereedYes
    StatusPublished
    ID Code55288

    Export

    Actions (login required)

    View Item

    Document Downloads

    More statistics for this item...