Gender bias in AI-based decision-making systems: a systematic literature review

Gender bias in AI-based decision-making systems: a systematic literature review
复制标题

基于人工智能的决策系统中的性别偏见:系统文献综述

DOI:
10.3127/ajis.v26i0.3835
复制
发表时间:
2022
期刊:
Australas. J. Inf. Syst.
影响因子:
--
通讯作者:
B. Abedin
B. Abedin
中科院分区:
--
文献类型:
--
作者:
Ayesha Nadeem;O. Marjanovic;B. Abedin

文献摘要

参考文献

被引文献

相似文献

相关文献和行业新闻表明,基于人工智能(AI)的决策系统可能会偏向性别,这反过来会影响个人和社会。信息系统(IS)领域已经认识到基于人工智能的成果及其影响的丰富贡献;然而,缺乏对基于人工智能的决策系统中性别偏见及其不利影响的管理的IS研究。因此,对基于人工智能的决策系统中性别偏见的日益关注正在引起人们的关注。特别是,有必要更好地了解促成因素和有效方法,以减轻基于人工智能的决策系统中的性别偏见。因此,本研究通过对现有文献进行系统性文献综述(SLR),并为基于人工智能的决策系统中的性别偏见管理提供了一个理论框架,从而为现有文献做出了贡献。SLR的结果表明,基于人工智能的决策系统中的性别偏见研究还没有很好地建立起来,突出了未来在这一领域的研究的巨大潜力。基于这一综述,我们将基于人工智能的决策系统中的性别偏见概念化为一个社会技术问题,并提出了一个理论框架,该框架提供了技术、组织和社会方法的组合,以及四个可能减轻偏见影响的命题。最后,本文考虑了未来在组织背景下基于人工智能的决策系统中管理性别偏见的研究。
The related literature and industry press suggest that artificial intelligence (AI)-based decision-making systems may be biased towards gender, which in turn impacts individuals and societies. The information system (IS) field has recognised the rich contribution of AI-based outcomes and their effects; however, there is a lack of IS research on the management of gender bias in AI-based decision-making systems and its adverse effects. Hence, the rising concern about gender bias in AI-based decision-making systems is gaining attention. In particular, there is a need for a better understanding of contributing factors and effective approaches to mitigating gender bias in AI-based decision-making systems. Therefore, this study contributes to the existing literature by conducting a Systematic Literature Review (SLR) of the extant literature and presenting a theoretical framework for the management of gender bias in AI-based decision-making systems. The SLR results indicate that the research on gender bias in AI-based decision-making systems is not yet well established, highlighting the great potential for future IS research in this area, as articulated in the paper. Based on this review, we conceptualise gender bias in AI-based decision-making systems as a socio-technical problem and propose a theoretical framework that offers a combination of technological, organisational, and societal approaches as well as four propositions to possibly mitigate the biased effects. Lastly, this paper considers future research on the management of gender bias in AI-based decision-making systems in the organisational context.
从软分类器到硬决策:我们能做到多公平?
DOI: 10.1145/3287560.3287561
发表时间: 2019
期刊: and Transparency - FAT* '19
影响因子: --
作者:
Canetti, Ran;Cohen, Aloni;Dikkala, Nishanth;Ramnarayan, Govind;Scheffler, Sarah;Smith, Adam
通讯作者: Smith, Adam
信息系统中的人工智能:最新技术和研究路线图
DOI: 10.17705/1cais.05017
发表时间: 2022
影响因子: 2
作者:
Ågerfalk, Pär J.;Conboy, Kieran;Crowston, Kevin;Eriksson Lundström, Jenny S.;Jarvenpaa, Sirkka;Ram, Sudha;Mikalef, Patrick
通讯作者: Mikalef, Patrick
DOI: 10.1007/s10551-018-3921-3
发表时间: 2019-12-01
影响因子: 6.1
作者:
Martin, Kirsten
通讯作者: Martin, Kirsten
DOI: 10.1109/tvcg.2019.2934262
发表时间: 2020-01-01
影响因子: 5.2
作者:
Ahn, Yongsu;Lin, Yu-Ru
通讯作者: Lin, Yu-Ru
DOI: 10.18653/v1/p19-1159
发表时间: 2019-06
期刊: --
影响因子: --
作者:
Tony Sun;Andrew Gaut;Shirlyn Tang;Yuxin Huang;Mai Elsherief;Jieyu Zhao;Diba Mirza;E. Belding-Royer;Kai-Wei Chang;William Yang Wang
通讯作者: Tony Sun;Andrew Gaut;Shirlyn Tang;Yuxin Huang;Mai Elsherief;Jieyu Zhao;Diba Mirza;E. Belding-Royer;Kai-Wei Chang;William Yang Wang