EUVIMEDEUVIMED entdecken
Crossref

FAIRNESS AND BIAS MITIGATION IN AI-BASED CREDIT SCORING USING ALTERNATIVE DATA: A FRAMEWORK FOR ETHICAL FINANCIAL INCLUSION

Derrick Atuobi Oware, Samuel Amfo Junior

EPRA International Journal of Multidisciplinary Research (IJMR) · 31.07.2025 · 1095-1101

DOI: 10.36713/epra23347 ↗

Kurz gefasst

Fachartikel in der Zeitschrift EPRA International Journal of Multidisciplinary Research (IJMR), online erschienen am 31.07.2025. Verfasst von Derrick Atuobi Oware und 1 weiteren Autor:innen. Die Arbeit wurde bisher 3-mal zitiert (Crossref).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

There are a lot of opportunities to speed up financial inclusion through the use of artificial intelligence (AI) and alternative data in credit scoring, especially for underprivileged groups that have been shut out of formal financial services. Systemic exclusion is reinforced by the inability of traditional credit scoring methods to account for the financial conduct of those without prior credit histories. The paper describes how more comprehensive, data-driven credit assessment models can be constructed by leveraging alternative data sources, such as web footprints, utility bill payments, and mobile phone usage patterns. However, there are also technical and ethical issues with using AI in this way, like algorithmic bias, opacity, and data privacy issues. This paper responds by offering a strategy for achieving equity and mitigating bias in AI-based credit assessment models using alternative data. The framework consists of four integral parts: responsible data gathering, bias-aware model development, fairness metrics, and regulatory conformity processes. Relying on case studies and model simulations, the paper explores how these parts cooperate to generate credit decisions and addresses how to enhance transparency, accountability, and trust. Specific attention is drawn to the necessity for inclusive data governance, stakeholder engagement, and the application of explainable AI techniques. By addressing both the promise and risks of this emerging field, the proposed framework contributes to ongoing work to synchronize technological innovation with ethical principles and promote equitable access to credit. Keywords: Financial Inclusion, Alternative Data, Credit Scoring, Algorithmic Fairness, Bias Mitigation

Zitieren

APA

Oware, D. A., & Junior, S. A. (2025). FAIRNESS AND BIAS MITIGATION IN AI-BASED CREDIT SCORING USING ALTERNATIVE DATA: A FRAMEWORK FOR ETHICAL FINANCIAL INCLUSION. EPRA International Journal of Multidisciplinary Research (IJMR), 1095-1101. https://doi.org/10.36713/epra23347

Vancouver

Oware DA, Junior SA. FAIRNESS AND BIAS MITIGATION IN AI-BASED CREDIT SCORING USING ALTERNATIVE DATA: A FRAMEWORK FOR ETHICAL FINANCIAL INCLUSION. EPRA International Journal of Multidisciplinary Research (IJMR). 2025:1095-1101. doi:10.36713/epra23347

BibTeX
@article{Oware2025,
  author = {Oware, Derrick Atuobi and Junior, Samuel Amfo},
  title = {FAIRNESS AND BIAS MITIGATION IN AI-BASED CREDIT SCORING USING ALTERNATIVE DATA: A FRAMEWORK FOR ETHICAL FINANCIAL INCLUSION},
  journal = {EPRA International Journal of Multidisciplinary Research (IJMR)},
  year = {2025},
  pages = {1095--1101},
  doi = {10.36713/epra23347}
}
Infokarte | EUVIMED

Artikeldaten derzeit nicht verfügbar.

Recherche