Evaluating Algorithmic Fairness in Machine Learning-Based Recruitment and Candidate Screening Systems

Siti Farah Abdullah

Evaluating Algorithmic Fairness in Machine Learning-Based Recruitment and Candidate Screening Systems

Author(s)	Siti Farah Abdullah
Country	Malaysia
Abstract	Machine learning is increasingly considered for recruitment and candidate-screening tasks because automated systems can process large numbers of applications and identify patterns associated with predefined job-related criteria. However, historical recruitment data may contain unequal selection patterns, measurement limitations, or proxy variables that influence automated predictions. Algorithmic fairness therefore represents an important methodological concern when machine learning is used to rank, classify, or prioritize job applicants. Predictive accuracy alone cannot determine whether a screening system produces equitable outcomes across relevant applicant groups. This hypothetical methodological study examines algorithmic fairness in machine learning-based recruitment and candidate-screening systems. The proposed framework separates predictive performance from group-level fairness assessment and compares three conceptual models: logistic regression, random forest classification, and a fairness-aware classification model. A fictional dataset containing 1,500 candidate applications is divided into hypothetical training, validation, and testing subsets. Fairness is evaluated using selection-rate difference, selection-rate ratio, and true-positive-rate difference across two anonymized applicant groups. The illustrative analysis shows that the highest predictive accuracy and the smallest group disparities do not necessarily occur in the same model configuration. Under the assigned hypothetical values, the random forest produces the strongest conventional predictive accuracy, whereas the fairness-aware model substantially reduces selection-rate and true-positive-rate differences while maintaining comparable predictive performance. These numerical outcomes are illustrative rather than experimentally observed results. The study concludes that fairness evaluation in automated recruitment should use multiple performance and equity measures, clear documentation of applicant features, subgroup-specific error analysis, and human review. No individual fairness metric can independently demonstrate that a recruitment process is legally, ethically, or practically fair. Empirical validation using representative data and appropriate organizational governance is therefore required before automated candidate-screening systems are used for consequential employment decisions.
Keywords	Algorithmic Fairness, Machine Learning, Recruitment Analytics, Candidate Screening, Discrimination-Aware Classification, Fairness Metrics, Automated Decision-Making, Employment Analytics, Responsible Artificial Intelligence, Human Resource Analytics.
Field	Engineering
Published In	Volume 1, Issue 5, November-December 2025
Published On	2025-11-27

View / Download PDF File

E-ISSN: 3067-5073

doi

CrossRef DOI is assigned to each research paper published in our journal.

IJMLAIDSE DOI prefix is
10.00000/IJMLAIDSE

Downloads

Research Paper Format Copyright Permission and Undertaking Form

All research papers published on this website are licensed under Creative Commons Attribution-ShareAlike 4.0 International License, and all rights belong to their respective authors/researchers.

CC-BY-SA

About IJMLAIDSE Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us	+1-872-302-3851	editor@uscse.com

International Journal of Machine Learning, AI & Data Science Evolution

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Monthly Scholarly International Journal

Evaluating Algorithmic Fairness in Machine Learning-Based Recruitment and Candidate Screening Systems

Share this