Debiasing Education Algorithms
Дебайасинг алгоритмов в образовании
2024-01-04
SCID: 54.1/5kdeynxd
Discuss with AI
ABROCAadversarial learningalgorithmic fairnesseducational settingsstudent dropout prediction
Figures from the paper
Abstract (AI)
This systematic literature review investigates the fairness of machine learning algorithms in educational settings, focusing on recent studies and their proposed solutions to address biases. Applications analyzed include student dropout prediction, performance prediction, forum post classification, and recommender systems. We identify common strategies, such as adjusting sample weights, bias attenuation methods, fairness through un/awareness, and adversarial learning. Commonly used metrics for fairness assessment include ABROCA, group difference in performance, and disparity metrics. The review underscores the need for context-specific approaches to ensure equitable treatment and reveals that most studies found no strict tradeoff between fairness and accuracy. We recommend evaluating fairness of data and features before algorithmic fairness to prevent algorithms from receiving discriminatory inputs, expanding the scope of education fairness studies beyond gender and race to include other demographic attributes, and assessing the impact of fair algorithms on end users, as human perceptions may not align with algorithmic fairness measures.
Key Findings
1
Common debiasing strategies include sample-weight adjustment, bias attenuation, fairness through awareness or unawareness, and adversarial learning.
2
Educational algorithm fairness is commonly evaluated using ABROCA, group performance differences, and disparity metrics.
3
Most reviewed studies reported no strict tradeoff between fairness and predictive accuracy, although approaches require contextual adaptation.
4
The review examines fairness interventions across educational applications including dropout prediction, performance prediction, forum classification, and recommender systems.
5
The review recommends auditing data and features before algorithms, studying demographic attributes beyond gender and race, and evaluating effects on end users.
Research Object
Machine learning algorithms used in educational settings, including student dropout and performance prediction, forum post classification, and recommender systems
Research Subject
Fairness, bias, and fairness–accuracy trade-offs in educational machine learning algorithms, including strategies and metrics for mitigating and assessing discriminatory outcomes
Publication Details
Publication Date
2024-01-04
Journal
Publisher
ISSN
Cited by
52
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest