Constructing categories: Moving beyond protected classes in algorithmic fairness

Constructing categories: Moving beyond protected classes in algorithmic fairness
复制标题

构建类别:在算法公平性方面超越受保护类别

DOI:
10.1002/asi.24643
复制
发表时间:
2022
影响因子:
3.5
通讯作者:
Bosch, Nigel
Bosch, Nigel
中科院分区:
管理学3区
文献类型:
--
作者:
Belitz, Clara;Ocumpaugh, Jaclyn;Ritter, Steven;Baker, Ryan S.;Fancsali, Stephen E.;Bosch, Nigel

文献摘要

参考文献

被引文献

相似文献

自动化、数据驱动的决策制定在各种应用领域越来越普遍。例如,在教育软件中,机器学习已被应用于选择学生要完成的下一个练习等任务。然而,机器学习方法并不总是对所有学生群体都同样有效。目前设计公平算法的方法倾向于关注关于受法律保护的类别(如种族或性别)的一小部分的统计措施。然而,仅仅关注受法律保护的类别可能会因为忽视身份的复杂性而限制我们对偏见和不公平的理解。我们提出了另一种方法来分类,在测量身份的社会学技术接地。通过从被研究的人群中收集调查数据和访谈,我们可以自下而上建立特定于背景的类别。然后,新兴类别可以与现存的算法公平策略相结合,以发现哪些身份群体没有得到很好的服务,从而发现应该改进或完全避免的算法。我们专注于教育应用程序,但目前的论点,这种方法应该更广泛地采用算法的公平性问题,在各种应用程序。
Automated, data‐driven decision making is increasingly common in a variety of application domains. In educational software, for example, machine learning has been applied to tasks like selecting the next exercise for students to complete. Machine learning methods, however, are not always equally effective for all groups of students. Current approaches to designing fair algorithms tend to focus on statistical measures concerning a small subset of legally protected categories like race or gender. Focusing solely on legally protected categories, however, can limit our understanding of bias and unfairness by ignoring the complexities of identity. We propose an alternative approach to categorization, grounded in sociological techniques of measuring identity. By soliciting survey data and interviews from the population being studied, we can build context‐specific categories from the bottom up. The emergent categories can then be combined with extant algorithmic fairness strategies to discover which identity groups are not well‐served, and thus where algorithms should be improved or avoided altogether. We focus on educational applications but present arguments that this approach should be adopted more broadly for issues of algorithmic fairness across a variety of applications.
种族和 SAT 成绩之间日益增强的相关性:来自加利福尼亚州的新发现
DOI: --
发表时间: 2015
期刊:
影响因子: --
作者:
Saul Geiser
通讯作者: Saul Geiser
DOI: --
发表时间: 2016
期刊:
影响因子: --
作者:
Brittany A. Aronson;Judson C. Laughter
通讯作者: Judson C. Laughter
教育数据挖掘模型的总体有效性:情感检测的案例研究
DOI: 10.1111/bjet.12156
发表时间: 2014
期刊: Br. J. Educ. Technol.
影响因子: --
作者:
Jaclyn L. Ocumpaugh;R. Baker;S. M. Gowda;N. Heffernan;Cristina Heffernan
通讯作者: Cristina Heffernan
DOI: --
发表时间: 2018
期刊: International Conference on Human Factors in Computing Systems
影响因子: --
作者:
Allison Woodruff;Sarah E. Fox;Steven Rousso;J. Warshaw
通讯作者: J. Warshaw
教育中的算法公平
DOI: --
发表时间: 2020
期刊: arXiv.org
影响因子: --
作者:
René F. Kizilcec;Hansol Lee
通讯作者: Hansol Lee