A general methodology to quantify biases in natural language data

Jiawei Chen, Anbang Xu, Zhe Liu, Yufan Guo, Xiaotong Liu, Yingbei Tong, Rama Akkiraju, John M. Carroll

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

Abstract

Biases in data, such as gender and racial stereotypes, are propagated through intelligent systems and amplified at end-user applications. Existing studies detect and quantify biases based on pre-defined attributes. However, in real practices, it is difficult to gather a comprehensive list of sensitive concepts for various categories of biases. We propose a general methodology to quantify dataset biases by measuring the difference of its data distribution with a reference dataset using Maximum Mean Discrepancy. For the case of natural language data, we show that lexicon-based features quantify explicit stereotypes, while deep learning-based features further capture implicit stereotypes represented by complex semantics. Our method provides a more flexible way to detect potential biases.

Original languageEnglish (US)
Title of host publicationCHI EA 2020 - Extended Abstracts of the 2020 CHI Conference on Human Factors in Computing Systems
PublisherAssociation for Computing Machinery
ISBN (Electronic)9781450368193
DOIs
StatePublished - Apr 25 2020
Event2020 ACM CHI Conference on Human Factors in Computing Systems, CHI EA 2020 - Honolulu, United States
Duration: Apr 25 2020Apr 30 2020

Publication series

NameConference on Human Factors in Computing Systems - Proceedings

Conference

Conference2020 ACM CHI Conference on Human Factors in Computing Systems, CHI EA 2020
CountryUnited States
CityHonolulu
Period4/25/204/30/20

All Science Journal Classification (ASJC) codes

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

Fingerprint Dive into the research topics of 'A general methodology to quantify biases in natural language data'. Together they form a unique fingerprint.

Cite this