Slide to Explore 'What If': An Analysis of Explainable Interfaces

Louie Søs Meyer; Jichen Zhu

doi:10.1145/3677045.3685416

Slide to Explore 'What If': An Analysis of Explainable Interfaces

Human-AI Interaction Lab

Research output: Journal Article or Conference Article in Journal › Conference article › Research › peer-review

Abstract

Most users interact with explainable AI (XAI) systems through an interface. Despite the growing assemblage of human-centered XAI literature, there is still limited research that investigates how explanations are presented in interfaces. With this preliminary study, we propose a novel approach to analyze the content and interactions of explainable interfaces by mapping UI elements and interactions to the type of explanation they convey. Based on analysis of 28 explainable interfaces, we offer initial findings of how various interaction concepts and UI elements are used to provide different explanations.

Original language	English
Article number	2
Conference proceedings	Proceedings of NordiCHI
Pages (from-to)	1-6
Number of pages	6
DOIs	https://doi.org/10.1145/3677045.3685416
Publication status	Published - 13 Oct 2024
Event	Nordic Conference on Human-Computer Interaction - Uppsala, Sweden Duration: 13 Oct 2024 → 16 Oct 2024 https://dl.acm.org/doi/proceedings/10.1145/3677045

Conference

Conference	Nordic Conference on Human-Computer Interaction
Country/Territory	Sweden
City	Uppsala
Period	13/10/2024 → 16/10/2024
Internet address	https://dl.acm.org/doi/proceedings/10.1145/3677045

Keywords

explainable AI interfaces
user interface elements
explanation types
interaction concepts
human-centered XAI

Access to Document

10.1145/3677045.3685416

ExPAI: Explainable Personalization AI for Health and Well-Being
Zhu, J. (PI), Malmborg, L. (CoI), Meyer, L. S. (CoI) & Maden, W. L. A. V. D. (CoI)
Novo Nordisk Foundation
01/01/2021 → 01/05/2027
Project: Research

Cite this

@inproceedings{e8c86569669140c88ebdb350b2457a54,

title = "Slide to Explore 'What If': An Analysis of Explainable Interfaces",

abstract = "Most users interact with explainable AI (XAI) systems through an interface. Despite the growing assemblage of human-centered XAI literature, there is still limited research that investigates how explanations are presented in interfaces. With this preliminary study, we propose a novel approach to analyze the content and interactions of explainable interfaces by mapping UI elements and interactions to the type of explanation they convey. Based on analysis of 28 explainable interfaces, we offer initial findings of how various interaction concepts and UI elements are used to provide different explanations.",

keywords = "explainable AI interfaces, user interface elements, explanation types, interaction concepts, human-centered XAI",

author = "Meyer, \{Louie S{\o}s\} and Jichen Zhu",

year = "2024",

month = oct,

day = "13",

doi = "10.1145/3677045.3685416",

language = "English",

pages = "1--6",

journal = "Proceedings of NordiCHI",

publisher = "Association for Computing Machinery",

note = "Nordic Conference on Human-Computer Interaction, NordiCHI ; Conference date: 13-10-2024 Through 16-10-2024",

url = "https://dl.acm.org/doi/proceedings/10.1145/3677045",

}

TY - GEN

T1 - Slide to Explore 'What If': An Analysis of Explainable Interfaces

AU - Meyer, Louie Søs

AU - Zhu, Jichen

PY - 2024/10/13

Y1 - 2024/10/13

N2 - Most users interact with explainable AI (XAI) systems through an interface. Despite the growing assemblage of human-centered XAI literature, there is still limited research that investigates how explanations are presented in interfaces. With this preliminary study, we propose a novel approach to analyze the content and interactions of explainable interfaces by mapping UI elements and interactions to the type of explanation they convey. Based on analysis of 28 explainable interfaces, we offer initial findings of how various interaction concepts and UI elements are used to provide different explanations.

AB - Most users interact with explainable AI (XAI) systems through an interface. Despite the growing assemblage of human-centered XAI literature, there is still limited research that investigates how explanations are presented in interfaces. With this preliminary study, we propose a novel approach to analyze the content and interactions of explainable interfaces by mapping UI elements and interactions to the type of explanation they convey. Based on analysis of 28 explainable interfaces, we offer initial findings of how various interaction concepts and UI elements are used to provide different explanations.

KW - explainable AI interfaces

KW - user interface elements

KW - explanation types

KW - interaction concepts

KW - human-centered XAI

U2 - 10.1145/3677045.3685416

DO - 10.1145/3677045.3685416

M3 - Conference article

SP - 1

EP - 6

JO - Proceedings of NordiCHI

JF - Proceedings of NordiCHI

M1 - 2

T2 - Nordic Conference on Human-Computer Interaction

Y2 - 13 October 2024 through 16 October 2024

ER -

Slide to Explore 'What If': An Analysis of Explainable Interfaces

Abstract

Conference

Keywords

Access to Document

Fingerprint

Projects

ExPAI: Explainable Personalization AI for Health and Well-Being

Cite this