EUCA: A Practical Prototyping Framework towards End-User-Centered Explainable Artificial Intelligence
The ability to explain decisions to its end-users is a necessity to deploy AI as critical decision support. Yet making AI explainable to end-users is a relatively ignored and challenging problem. To bridge the gap, we first identified twelve end-user-friendly explanatory forms that do not require technical knowledge to comprehend, including feature-, example-, and rule-based explanations. We then instantiated the explanatory forms as prototyping cards in four AI-assisted critical decision-making tasks, and conducted a user study to co-design low-fidelity prototypes with 32 layperson participants. The results verified the relevance of using the explanatory forms as building blocks of explanations, and identified their proprieties (pros, cons, applicable explainability needs, and design implications). The explanatory forms, their proprieties, and prototyping support constitute the End-User-Centered explainable AI framework EUCA. It serves as a practical prototyping toolkit for HCI/AI practitioners and researchers to build end-user-centered explainable AI. The EUCA framework is available at http://weina.me/end-user-xai
READ FULL TEXT