Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Tell me more: Intent Fulfilment Framework for Enhancing User Experiences in Conversational XAI

Anjana Wijekoon, David Corsar, Nirmalie Wiratunga, Kyle Martin, Pedram Salimi

TL;DR

The paper addresses the challenge that explainable AI explanations must accommodate diverse user intents and support meaningful follow-up interactions. It introduces the Intent Fulfilment Framework (IFF), an ontology-guided typology linking user intents, questions, explanation types, and follow-up relations, and the Explanation Experience Dialogue Model (EEDM) that implements IFF in a conversational setting. A loan-approval use-case and co-design with industry partners ground the approach, and a comparative user study shows that enabling Explanation Followups improves user engagement, perceived utility, and overall experience. The work demonstrates that one explanation does not fit all and provides a scalable, reusable framework for designing user-centered, explainable interactions across domains.

Abstract

The evolution of Explainable Artificial Intelligence (XAI) has emphasised the significance of meeting diverse user needs. The approaches to identifying and addressing these needs must also advance, recognising that explanation experiences are subjective, user-centred processes that interact with users towards a better understanding of AI decision-making. This paper delves into the interrelations in multi-faceted XAI and examines how different types of explanations collaboratively meet users' XAI needs. We introduce the Intent Fulfilment Framework (IFF) for creating explanation experiences. The novelty of this paper lies in recognising the importance of "follow-up" on explanations for obtaining clarity, verification and/or substitution. Moreover, the Explanation Experience Dialogue Model integrates the IFF and "Explanation Followups" to provide users with a conversational interface for exploring their explanation needs, thereby creating explanation experiences. Quantitative and qualitative findings from our comparative user study demonstrate the impact of the IFF in improving user engagement, the utility of the AI system and the overall user experience. Overall, we reinforce the principle that "one explanation does not fit all" to create explanation experiences that guide the complex interaction through conversation.

Tell me more: Intent Fulfilment Framework for Enhancing User Experiences in Conversational XAI

TL;DR

The paper addresses the challenge that explainable AI explanations must accommodate diverse user intents and support meaningful follow-up interactions. It introduces the Intent Fulfilment Framework (IFF), an ontology-guided typology linking user intents, questions, explanation types, and follow-up relations, and the Explanation Experience Dialogue Model (EEDM) that implements IFF in a conversational setting. A loan-approval use-case and co-design with industry partners ground the approach, and a comparative user study shows that enabling Explanation Followups improves user engagement, perceived utility, and overall experience. The work demonstrates that one explanation does not fit all and provides a scalable, reusable framework for designing user-centered, explainable interactions across domains.

Abstract

The evolution of Explainable Artificial Intelligence (XAI) has emphasised the significance of meeting diverse user needs. The approaches to identifying and addressing these needs must also advance, recognising that explanation experiences are subjective, user-centred processes that interact with users towards a better understanding of AI decision-making. This paper delves into the interrelations in multi-faceted XAI and examines how different types of explanations collaboratively meet users' XAI needs. We introduce the Intent Fulfilment Framework (IFF) for creating explanation experiences. The novelty of this paper lies in recognising the importance of "follow-up" on explanations for obtaining clarity, verification and/or substitution. Moreover, the Explanation Experience Dialogue Model integrates the IFF and "Explanation Followups" to provide users with a conversational interface for exploring their explanation needs, thereby creating explanation experiences. Quantitative and qualitative findings from our comparative user study demonstrate the impact of the IFF in improving user engagement, the utility of the AI system and the overall user experience. Overall, we reinforce the principle that "one explanation does not fit all" to create explanation experiences that guide the complex interaction through conversation.
Paper Structure (31 sections, 2 equations, 15 figures)

This paper contains 31 sections, 2 equations, 15 figures.

Table of Contents

  1. Introduction
  2. Related Work
  3. Multi-faceted XAI
  4. Conversational XAI
  5. Intent Fulfilment
  6. Ontology
  7. Intent Fulfilment Framework
  8. Explanation Followup Relations
  9. IFF Development Methodology
  10. Explanation Experience Dialogue Model
  11. Implementation
  12. Evaluation
  13. Study Application
  14. Recruitment
  15. Analysis and Results
  16. ...and 16 more sections

Figures (15)

  • Figure 1: Intent Fulfilment Ontology
  • Figure 2: Intent Fulfilment Framework (IFF). The bold black borders indicate the IFF selected for Loan Applicants interacting with a Loan Approval AI system, used as the study application in the user study.
  • Figure 3: Explanation Experience Dialogue Model
  • Figure 4: A sample conversation between a Loan Applicant user and the Chatbot annotated with dialogue topics, locutions and references to the IFF
  • Figure 5: Evaluation Questionnaire Response Difference of Group B over Group A (count B - count A)
  • ...and 10 more figures

Theorems & Definitions (5)

  • Definition 4.1: Explanation Dialogue Model
  • Definition 4.2: Explanation Experience Dialogue Model
  • Definition 4.3: Topic Layer
  • Definition 4.4: Dialogue Layer
  • Definition 4.5: Control Layer