Cardiff Metropolitan University
Browse
- No file added yet -

An Explainable AI Solution: Exploring Extended Reality as a Way to Make Artificial Intelligence More Transparent and Trustworthy

Download (3.79 MB)
conference contribution
posted on 2023-10-10, 09:12 authored by Richard Wheeler, Fiona CarrollFiona Carroll

 Machine Learning (ML) and Artificial Intelligence (AI) have not only transformed the way we work (i.e. how we integrate information, analyse data, and how we make decisions) but also how organisations operate (i.e. adding new business processes and services etc.). In fact, many private, public and even third sector organisations are now capitalising on the true value of having systems that can learn on their own without any human intervention. However, with these benefits also come challenges regarding project productivity and collaboration. In detail, the need to explain how these systems work and how organisations interpret their output to achieve transparency and trust. This paper details the potential of using Extended reality (XR) as a way for enabling Explainable AI (XAI) focusing on the design and development of a novel XAI XR solution. The paper also highlights the ‘positive’ responses from an initial solution evaluation study noting participant’s impressions of the solution. It then makes recommendations for further research and development into the effectiveness of XR for explainable AI. 

History

Presented at

International Conference on Cybersecurity, Situational Awareness and Social Media Cyber Science 2022; 20–21 June; Wales

Publisher

Springer

Version

  • AM (Accepted Manuscript)

Citation

Wheeler, R., Carroll, F. (2023) An Explainable AI Solution: Exploring Extended Reality as a Way to Make Artificial Intelligence More Transparent and Trustworthy. In: Onwubiko, C., et al. Proceedings of the International Conference on Cybersecurity, Situational Awareness and Social Media. Springer Proceedings in Complexity. Springer, Singapore. https://doi.org/10.1007/978-981-19-6414-5_15

Print ISSN

2213-8684

Electronic ISSN

2213-8692

ISBN

978-981-19-6414-5

Cardiff Met Affiliation

  • Cardiff School of Technologies

Cardiff Met Authors

Richard Wheeler Fiona Carroll

Copyright Holder

  • © The Authors

Publisher Rights Statement

https://www.springernature.com/gp/open-research/policies/accepted-manuscript-terms

Language

  • en

Usage metrics

    School of Technologies Research - Conference Papers

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC