English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Most existing evaluations of explainable machine learning (ML) methods relyon simplifying assumptions or proxies that do not reflect real-world use cases;the handful of more robust evaluations on real-world settings have shortcomingsin their design, resulting in limited conclusions of methods' real-worldutility. In this work, we seek to bridge this gap by conducting a study thatevaluates three popular explainable ML methods in a setting consistent with theintended deployment context. We build on a previous study on e-commerce frauddetection and make crucial modifications to its setup relaxing the simplifyingassumptions made in the original work that departed from the deploymentcontext. In doing so, we draw drastically different conclusions from theearlier work and find no evidence for the incremental utility of the testedmethods in the task. Our results highlight how seemingly trivial experimentaldesign choices can yield misleading conclusions, with lessons about thenecessity of not only evaluating explainable ML methods using tasks, data,users, and metrics grounded in the intended deployment contexts but alsodeveloping methods tailored to specific applications. In addition, we believethe design of this experiment can serve as a template for future study designsevaluating explainable ML methods in other real-world contexts.

On the Importance of Application-Grounded Experimental Design for Evaluating Explainable ML Methods