Word embeddings quantify 100 years of gender and ethnic stereotypes

Garg, Nikhil; Schiebinger, Londa; Jurafsky, Dan; Zou, James

doi:10.1073/pnas.1720347115

Cited by 836 publications

(978 citation statements)

References 38 publications

Supporting

Mentioning

870

Contrasting

Unclassified

Order By: Relevance

“…We also explore the stability of embeddings, but focus on a broader range of factors, and consider the effect of stability on downstream tasks. In contrast, Antoniak and Mimno focus on using word embeddings to analyze language (e.g., Garg et al, 2018), rather than to perform tasks.…”

Section: Related Workmentioning

confidence: 99%

Factors Influencing the Surprising Instability of Word Embeddings

Wendlandt¹,

Kummerfeld²,

Mihalcea³

2018

Proceedings of the 2018 Conference of the North American Chapter Of the Association for Computational Linguistics: Hu

101

View full text Add to dashboard Cite

Despite the recent popularity of word embedding methods, there is only a small body of work exploring the limitations of these representations. In this paper, we consider one aspect of embedding spaces, namely their stability. We show that even relatively high frequency words (100-200 occurrences) are often unstable. We provide empirical evidence for how various factors contribute to the stability of word embeddings, and we analyze the effects of stability on downstream tasks.

show abstract

Section: Related Workmentioning

confidence: 99%

Factors Influencing the Surprising Instability of Word Embeddings

Wendlandt¹,

Kummerfeld²,

Mihalcea³

2018

Proceedings of the 2018 Conference of the North American Chapter Of the Association for Computational Linguistics: Hu

101

View full text Add to dashboard Cite

show abstract

“…Natural language processing (NLP) algorithms have been reported to incorporate inherent bias when trained on human language . NLP techniques such as word embedding are now used to objectively evaluate gender and ethnic stereotypes in text data . In recent years, there have been unfortunate examples of nonmedical NLP and machine learning algorithms that have produced biased recommendations .…”

Section: Introductionmentioning

confidence: 99%

“…1,2 NLP techniques such as word embedding are now used to objectively evaluate gender and ethnic stereotypes in text data. 3 In recent years, there have been unfortunate examples of nonmedical NLP and machine learning algorithms that have produced biased recommendations. 4 These setbacks risk jeopardizing physician trust in machine learning-based clinical decision support tools.…”

Section: Introductionmentioning

confidence: 99%

Investigation of bias in an epilepsy machine learning algorithm trained on physician notes

et al. 2019

View full text Add to dashboard Cite

Racial disparities in the utilization of epilepsy surgery are well documented, but it is unknown whether a natural language processing (NLP) algorithm trained on physician notes would produce biased recommendations for epilepsy presurgical evaluations. To assess this, an NLP algorithm was trained to identify potential surgical candidates using 1097 notes from 175 epilepsy patients with a history of resective epilepsy surgery and 268 patients who achieved seizure freedom without surgery (total N = 443 patients). The model was tested on 8340 notes from 3776 patients with epilepsy whose surgical candidacy status was unknown (2029 male, 1747 female, median age = 9 years; age range = 0‐60 years). Multiple linear regression using demographic variables as covariates was used to test for correlations between patient race and surgical candidacy scores. After accounting for other demographic and socioeconomic variables, patient race, gender, and primary language did not influence surgical candidacy scores (P > .35 for all). Higher scores were given to patients >18 years old who traveled farther to receive care, and those who had a higher family income and public insurance (P < .001, .001, .001, and .01, respectively). Demographic effects on surgical candidacy scores appeared to reflect patterns in patient referrals.

show abstract

“…For example, studies have shown that-based on vectors learned from text corpora-computer programmer is more to male than female, and further found that this type of analogy reflects societal bias, which can be quantified through word vectors (Garg et al, 2018).…”

Section: Lsmentioning

confidence: 99%

The citation disadvantage of clinical research

Ke¹

2020

Journal of Informetrics

View full text Add to dashboard Cite

Biomedical research encompasses diverse types of activities, from basic science ("bench") to clinical medicine ("bedside") to bench-to-bedside translational research. It, however, remains unclear whether different types of research receive citations at varying rates. Here we aim to answer this question by using a newly proposed paper-level indicator that quantifies the extent to which a paper is basic science or clinical medicine.Applying this measure to 5 million biomedical papers, we find a systematic citation disadvantage of clinical oriented papers; they tend to garner far fewer citations and are less likely to be hit works than papers oriented towards basic science. At the same time, clinical research has a higher variance in its citation. We also find that the citation difference between basic and clinical research decreases, yet still persists, if longer citation-window is used. Given the increasing adoption of short-term, citation-based bibliometric indicators in funding decisions, the under-cited effect of clinical research may provide disincentives for bio-researchers to venture into the translation of basic scientific discoveries into clinical applications, thus providing explanations of reasons behind the existence of the gap between basic and clinical research that is commented as "valley of death" and the commentary of "extinction" risk of translational researchers. Our work may provide insights to policy-makers on how to evaluate different types of biomedical research.

show abstract

Word embeddings quantify 100 years of gender and ethnic stereotypes

Cited by 836 publications

References 38 publications

Factors Influencing the Surprising Instability of Word Embeddings

Factors Influencing the Surprising Instability of Word Embeddings

Investigation of bias in an epilepsy machine learning algorithm trained on physician notes

The citation disadvantage of clinical research

Contact Info

Product

Resources

About