University of Twente Student Theses
Improving trust-building through more transparent conversational agent communication, in the context of medical decision support.
Ali, Ali (2024) Improving trust-building through more transparent conversational agent communication, in the context of medical decision support.
|
PDF
1MB |
| Abstract: | There is a growing interest in utilising Large Language Models (LLMs) and Conversational Agents (CAs) in eHealth applications such as digital genetic consultancy. However, current problems such as hallucinatory outputs and low transparency negatively affect the trust that users put in such systems. Such problems require solutions before CAs could be safely incorporated in critical applications such as medical consultancy. In this paper we dis- cuss some of the recent efforts at improving the trust and transparency of CAs by examining recent advancements in Explainable AI (XAI) and how explanations are incorporated in the human-CA interaction in an easily un- derstandable format. Furthermore, we conduct an online survey to compare which of the two approaches of representing explanations to users (natural language explanations & UI indicators) result in more transparency and consequentially more realistic judgements of the CA and the information it generates. |
| Item Type: | Essay (Bachelor) |
| Faculty: | EEMCS: Electrical Engineering, Mathematics and Computer Science |
| Subject: | 05 communication studies, 54 computer science |
| Programme: | Business & IT BSc (56066) |
| Link to this item: | https://purl.utwente.nl/essays/101565 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page