Abstract
This paper presents a systematic literature review of algorithmic bias in decision-making from 2010 to 2023. Algorithmic bias in decision-making refers to the systematic and unfair outcomes produced by intelligent systems, which can arise from biased training data, flawed algorithm design, and socio-cultural contexts. The review identifies key definitions and sources of algorithmic bias, including data bias, design bias, and contextual bias, focusing on their impact on decision-making processes. It examines the dimensions of algorithmic bias, such as racial and gender biases, and highlights their implications in decision-making domains like marketing, hiring, and personalised pricing. The study underscores the critical need for transparency, ethical considerations, and human oversight to ensure fair and equitable decision-making by intelligent systems. Recommendations for future research and practical strategies for mitigating algorithmic bias are discussed, emphasising interdisciplinary approaches and comprehensive frameworks to address this pervasive issue in decision-making.
Recommended Citation
Alowayfi, Razan; Dennehy, Denis; and Dwivedi, Yogesh, "A Systematic Review of Algorithmic Bias in Decision-Making" (2024). SaudiCIS 2024 Proceedings. 49.
https://aisel.aisnet.org/saudicis2024/49