Harnessing value from data science in business: ensuring explainability and fairness of solutions

TL;DR

This paper explores how to ensure fairness and explainability in AI solutions for business, providing practical methods, algorithm audits, and future research directions to improve data-driven decision-making.

Contribution

It offers a comprehensive overview of fairness and explainability in AI, including bias mitigation techniques, explanation quality metrics, and application case studies.

Findings

Bias mitigation methods for fairness in AI

Techniques for quantifying explanation quality

Future research directions in XAI and fairness

Abstract

The paper introduces concepts of fairness and explainability (XAI) in artificial intelligence, oriented to solve a sophisticated business problems. For fairness, the authors discuss the bias-inducing specifics, as well as relevant mitigation methods, concluding with a set of recipes for introducing fairness in data-driven organizations. Additionally, for XAI, the authors audit specific algorithms paired with demonstrational business use-cases, discuss a plethora of techniques of explanations quality quantification and provide an overview of future research avenues.