Addressing age-related bias in sentiment analysis

Mark Díaz, Isaac Johnson, Amanda Lazar, Anne Marie Piper, Darren Gergle

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

Computational approaches to text analysis are useful in understanding aspects of online interaction, such as opinions and subjectivity in text. Yet, recent studies have identified various forms of bias in language-based models, raising concerns about the risk of propagating social biases against certain groups based on sociodemographic factors (e.g., gender, race, geography). In this study, we contribute a systematic examination of the application of language models to study discourse on aging. We analyze the treatment of age-related terms across 15 sentiment analysis models and 10 widely-used GloVe word embeddings and attempt to alleviate bias through a method of processing model training data. Our results demonstrate that significant age bias is encoded in the outputs of many sentiment analysis algorithms and word embeddings. We discuss the models' characteristics in relation to output bias and how these models might be best incorporated into research.

Original languageEnglish (US)
Title of host publicationCHI 2018 - Extended Abstracts of the 2018 CHI Conference on Human Factors in Computing Systems
Subtitle of host publicationEngage with CHI
PublisherAssociation for Computing Machinery
ISBN (Electronic)9781450356206, 9781450356213
DOIs
StatePublished - Apr 20 2018
Event2018 CHI Conference on Human Factors in Computing Systems, CHI 2018 - Montreal, Canada
Duration: Apr 21 2018Apr 26 2018

Publication series

NameConference on Human Factors in Computing Systems - Proceedings
Volume2018-April

Other

Other2018 CHI Conference on Human Factors in Computing Systems, CHI 2018
CountryCanada
CityMontreal
Period4/21/184/26/18

Fingerprint

Aging of materials
Processing

Keywords

  • Aging
  • Older adults, algorithmic bias
  • Sentiment analysis

ASJC Scopus subject areas

  • Human-Computer Interaction
  • Computer Graphics and Computer-Aided Design
  • Software

Cite this

Díaz, M., Johnson, I., Lazar, A., Piper, A. M., & Gergle, D. (2018). Addressing age-related bias in sentiment analysis. In CHI 2018 - Extended Abstracts of the 2018 CHI Conference on Human Factors in Computing Systems: Engage with CHI (Conference on Human Factors in Computing Systems - Proceedings; Vol. 2018-April). Association for Computing Machinery. https://doi.org/10.1145/3173574.3173986
Díaz, Mark ; Johnson, Isaac ; Lazar, Amanda ; Piper, Anne Marie ; Gergle, Darren. / Addressing age-related bias in sentiment analysis. CHI 2018 - Extended Abstracts of the 2018 CHI Conference on Human Factors in Computing Systems: Engage with CHI. Association for Computing Machinery, 2018. (Conference on Human Factors in Computing Systems - Proceedings).
@inproceedings{9e0e5ec9bf5f42749f9348eea201c5b9,
title = "Addressing age-related bias in sentiment analysis",
abstract = "Computational approaches to text analysis are useful in understanding aspects of online interaction, such as opinions and subjectivity in text. Yet, recent studies have identified various forms of bias in language-based models, raising concerns about the risk of propagating social biases against certain groups based on sociodemographic factors (e.g., gender, race, geography). In this study, we contribute a systematic examination of the application of language models to study discourse on aging. We analyze the treatment of age-related terms across 15 sentiment analysis models and 10 widely-used GloVe word embeddings and attempt to alleviate bias through a method of processing model training data. Our results demonstrate that significant age bias is encoded in the outputs of many sentiment analysis algorithms and word embeddings. We discuss the models' characteristics in relation to output bias and how these models might be best incorporated into research.",
keywords = "Aging, Older adults, algorithmic bias, Sentiment analysis",
author = "Mark D{\'i}az and Isaac Johnson and Amanda Lazar and Piper, {Anne Marie} and Darren Gergle",
year = "2018",
month = "4",
day = "20",
doi = "10.1145/3173574.3173986",
language = "English (US)",
series = "Conference on Human Factors in Computing Systems - Proceedings",
publisher = "Association for Computing Machinery",
booktitle = "CHI 2018 - Extended Abstracts of the 2018 CHI Conference on Human Factors in Computing Systems",

}

Díaz, M, Johnson, I, Lazar, A, Piper, AM & Gergle, D 2018, Addressing age-related bias in sentiment analysis. in CHI 2018 - Extended Abstracts of the 2018 CHI Conference on Human Factors in Computing Systems: Engage with CHI. Conference on Human Factors in Computing Systems - Proceedings, vol. 2018-April, Association for Computing Machinery, 2018 CHI Conference on Human Factors in Computing Systems, CHI 2018, Montreal, Canada, 4/21/18. https://doi.org/10.1145/3173574.3173986

Addressing age-related bias in sentiment analysis. / Díaz, Mark; Johnson, Isaac; Lazar, Amanda; Piper, Anne Marie; Gergle, Darren.

CHI 2018 - Extended Abstracts of the 2018 CHI Conference on Human Factors in Computing Systems: Engage with CHI. Association for Computing Machinery, 2018. (Conference on Human Factors in Computing Systems - Proceedings; Vol. 2018-April).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Addressing age-related bias in sentiment analysis

AU - Díaz, Mark

AU - Johnson, Isaac

AU - Lazar, Amanda

AU - Piper, Anne Marie

AU - Gergle, Darren

PY - 2018/4/20

Y1 - 2018/4/20

N2 - Computational approaches to text analysis are useful in understanding aspects of online interaction, such as opinions and subjectivity in text. Yet, recent studies have identified various forms of bias in language-based models, raising concerns about the risk of propagating social biases against certain groups based on sociodemographic factors (e.g., gender, race, geography). In this study, we contribute a systematic examination of the application of language models to study discourse on aging. We analyze the treatment of age-related terms across 15 sentiment analysis models and 10 widely-used GloVe word embeddings and attempt to alleviate bias through a method of processing model training data. Our results demonstrate that significant age bias is encoded in the outputs of many sentiment analysis algorithms and word embeddings. We discuss the models' characteristics in relation to output bias and how these models might be best incorporated into research.

AB - Computational approaches to text analysis are useful in understanding aspects of online interaction, such as opinions and subjectivity in text. Yet, recent studies have identified various forms of bias in language-based models, raising concerns about the risk of propagating social biases against certain groups based on sociodemographic factors (e.g., gender, race, geography). In this study, we contribute a systematic examination of the application of language models to study discourse on aging. We analyze the treatment of age-related terms across 15 sentiment analysis models and 10 widely-used GloVe word embeddings and attempt to alleviate bias through a method of processing model training data. Our results demonstrate that significant age bias is encoded in the outputs of many sentiment analysis algorithms and word embeddings. We discuss the models' characteristics in relation to output bias and how these models might be best incorporated into research.

KW - Aging

KW - Older adults, algorithmic bias

KW - Sentiment analysis

UR - http://www.scopus.com/inward/record.url?scp=85046977477&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85046977477&partnerID=8YFLogxK

U2 - 10.1145/3173574.3173986

DO - 10.1145/3173574.3173986

M3 - Conference contribution

T3 - Conference on Human Factors in Computing Systems - Proceedings

BT - CHI 2018 - Extended Abstracts of the 2018 CHI Conference on Human Factors in Computing Systems

PB - Association for Computing Machinery

ER -

Díaz M, Johnson I, Lazar A, Piper AM, Gergle D. Addressing age-related bias in sentiment analysis. In CHI 2018 - Extended Abstracts of the 2018 CHI Conference on Human Factors in Computing Systems: Engage with CHI. Association for Computing Machinery. 2018. (Conference on Human Factors in Computing Systems - Proceedings). https://doi.org/10.1145/3173574.3173986