An Explainable AI Framework for Customer Segmentation and Personalized Marketing Decision-Making in Online Retail

Abstract

Customer segmentation is a basis of personalized marketing, however, unsupervised clustering algorithms are fundamentally opaque: they cluster customers, but do not explain why a particular customer belongs in a particular cluster in interpretable terms. In this work, we propose and empirically validate an explainable AI (XAI) framework that combines unsupervised customer segmentation with post-hoc, surrogate-model-based explanation.