Evaluation of Quality Requirements for Explanations in AI-based Healthcare Systems

Zubaria Inayat*

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

28 Downloads (Pure)

Abstract

In the field of explainable artificial intelligence (XAI), methods are being developed to explain AI results. These methods form the range of implementation choices available to XAI designers when dealing with the explainability requirements to a system. While in the discipline of Requirements Engineering, explainability has been conceptualized and operationalized as a nonfunctional requirement, there was so far little focus specifically on the quality aspects of the explanations themselves. Yet, quality requirements issues pertaining to the explanations of AI systems lead to issues such as lack of transparency, trust, and user confidence. The present PhD research makes a step towards closing this gap. The research aims to formulate a solution for determining the quality of explanations in AI systems, particularly in the healthcare domain. We believe that this research will benefit healthcare professionals in maintaining confidence and trust in AI-based healthcare systems.

Original languageEnglish
Title of host publicationJoint Proceedings of REFSQ-2023 Workshops, Doctoral Symposium, Posters & Tools Track, and Journal Early Feedback Track
Subtitle of host publicationCo-located with REFSQ 2023. Barcelona, Catalunya, Spain, April 17, 2023
EditorsAlessio Ferrari, Birgit Penzenstadler, Irit Hadar, Shola Oyedeji, Sallam Abualhaija, Renata Guizzardi
PublisherCEUR
Number of pages7
Publication statusPublished - 2023
Event29th International Working Conference on Requirement Engineering, REFSQ 2023 - Barcelona, Spain
Duration: 17 Apr 202320 Apr 2023
Conference number: 29

Publication series

NameCEUR workshop proceedings
PublisherRheinisch Westfälische Technische Hochschule
Volume3378
ISSN (Print)1613-0073

Conference

Conference29th International Working Conference on Requirement Engineering, REFSQ 2023
Abbreviated titleREFSQ 2023
Country/TerritorySpain
CityBarcelona
Period17/04/2320/04/23

Keywords

  • Artificial Intelligence in medicine
  • Empirical research method`
  • Explainable artificial intelligence
  • Healthcare
  • Quality requirements
  • Requirements for explanations

Fingerprint

Dive into the research topics of 'Evaluation of Quality Requirements for Explanations in AI-based Healthcare Systems'. Together they form a unique fingerprint.

Cite this