Explainable AI-Driven Recommendation Systems and Consumer Buying Behaviour: The Mediating Role of Trust and Perceived Personalization in E-Commerce

Abstract

The rapid diffusion of artificial intelligence (AI) in e-commerce hastransformed consumer interactions with digital platforms through algorithmicrecommendation systems. However, their complexity and opacity raiseconcerns about transparency and trust. This study examines how explainableAI-driven recommendation quality influences purchase intention, focusing onperceived personalization and consumer trust. Using a quantitative, cross-sectional design, data were collected from 99 active e-commerce usersthrough a structured questionnaire and analysed using partial least squaresstructural equation modelling (PLS-SEM) and bootstrapping. Themeasurement model demonstrated satisfactory reliability, convergentvalidity, and discriminant validity. Results show that explainable AI-drivenrecommendation quality significantly influences perceived personalization (β= 0.434, p < 0.001) and consumer trust (β = 0.297, p = 0.002). Perceivedpersonalization significantly enhances consumer trust (β = 0.371, p < 0.001),while consumer trust positively affects purchase intention (β = 0.588, p <0.001). Significant indirect effects were also identified, including a serialindirect effect through perceived personalization and consumer trust (β =0.095, p = 0.002). The findings highlight explainability as an importantmechanism shaping consumer perceptions, trust, and purchase intention in e-commerce.

Keywords: Explainable Artificial Intelligence (XAI), AI-DrivenRecommendation Systems, Perceived Personalization, Consumer Trust,Purchase Intention, E-Commerce, PLS-SEM