Trade-offs Recommender Systems Make

How are recommender systems biased against minorities? Code review recommenders are designed to get work done faster and better. It would be interesting to see how our quest for speed and quality drives opportunities away. I have linked a few articles substantiating the field, the long-term consequences, and initial studies to understand the fairness landscape of code reviewer recommendations.

Good Reads

[1] Fairness Analysis of Machine Learning-Based Code Reviewer Recommendation https://link-springer-com.proxy-ub.rug.nl/chapter/10.1007/978-3-031-71975-2_4

[2] Evaluating unfairness of popularity bias in recommender systems: A comprehensive user-centric analysis https://www-sciencedirect-com.proxy-ub.rug.nl/science/article/pii/S0306457322002011

[3] Longitudinal Impact of Preference Biases on Recommender Systems’ Performance https://pubsonline.informs.org/doi/abs/10.1287/isre.2021.0133

[4] A review of code reviewer recommendation studies: Challenges and future directions https://www-sciencedirect-com.proxy-ub.rug.nl/science/article/pii/S0167642321000459

[5] Mitigating turnover with code review recommendation: balancing expertise, workload, and knowledge distribution https://dl-acm-org.proxy-ub.rug.nl/doi/abs/10.1145/3377811.3380335

[6] Does Reviewer Recommendation Help Developers?

[7] CoRReCT: code reviewer recommendation in GitHub based on cross-project and technology experience https://dl-acm-org.proxy-ub.rug.nl/doi/abs/10.1145/2889160.2889244