Towards a Novel Measure of User Trust in XAI Systems

TL;DR

This paper introduces a new trust measure for XAI systems that combines performance and trust indicators, validated through case studies showing improved sensitivity and refinement over existing methods.

Contribution

The paper proposes a novel, objective trust metric for XAI systems that enhances their evaluation and refinement capabilities.

Findings

Improved trust measurement sensitivity in case studies

Enhanced system refinement using the new metric

Outperforms existing trust evaluation methods

Abstract

The increasing reliance on Deep Learning models, combined with their inherent lack of transparency, has spurred the development of a novel field of study known as eXplainable AI (XAI) methods. These methods seek to enhance the trust of end-users in automated systems by providing insights into the rationale behind their decisions. This paper presents a novel trust measure in XAI systems, allowing their refinement. Our proposed metric combines both performance metrics and trust indicators from an objective perspective. To validate this novel methodology, we conducted three case studies showing an improvement respect the state-of-the-art, with an increased sensitiviy to different scenarios.