Qualitative Analysis of Depression Models by Demographics

Carlos Aguirre, Mark Dredze

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

Abstract

Models for identifying depression using social media text exhibit biases towards different gender and racial/ethnic groups. Factors like representation and balance of groups within the dataset are contributory factors, but difference in content and social media use may further explain these biases. We present an analysis of the content of social media posts from different demographic groups. Our analysis shows that there are content differences between depression and control subgroups across demographic groups, and that temporal topics and demographic-specific topics are correlated with downstream depression model error. We discuss the implications of our work on creating future datasets, as well as designing and training models for mental health.

Original languageEnglish (US)
Title of host publicationComputational Linguistics and Clinical Psychology
Subtitle of host publicationImproving Access, CLPsych 2021 - Proceedings of the 7th Workshop, in conjunction with NAACL 2021
EditorsNazli Goharian, Philip Resnik, Andrew Yates, Molly Ireland, Kate Niederhoffer, Rebecca Resnik
PublisherAssociation for Computational Linguistics (ACL)
Pages169-180
Number of pages12
ISBN (Electronic)9781954085411
StatePublished - 2021
Event7th Workshop on Computational Linguistics and Clinical Psychology: Improving Access, CLPsych 2021 - Virtual, Online
Duration: Jun 11 2021 → …

Publication series

NameComputational Linguistics and Clinical Psychology: Improving Access, CLPsych 2021 - Proceedings of the 7th Workshop, in conjunction with NAACL 2021

Conference

Conference7th Workshop on Computational Linguistics and Clinical Psychology: Improving Access, CLPsych 2021
CityVirtual, Online
Period6/11/21 → …

ASJC Scopus subject areas

  • Language and Linguistics
  • Computer Networks and Communications
  • Speech and Hearing

Fingerprint

Dive into the research topics of 'Qualitative Analysis of Depression Models by Demographics'. Together they form a unique fingerprint.

Cite this