Chatbot Research and Design : 6th International Workshop, CONVERSATIONS 2022, Amsterdam, The Netherlands, November 22–23, 2022, Revised Selected Papers

The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Preface 6 Organization 10 Contents 12 Chatbot Users and User Experience 14 Hi, I’m Cecil(y) the Smoking Cessation Chatbot: The Effectiveness of Motivational Interviewing and Confrontational Counseling Chatbots and the Moderating Role of the Need for Autonomy and Self-efficacy 15 1 Introduction 15 2 Conceptual Framework 17 2.1 Communication Styles for Smoking Cessation Chatbots 17 2.2 Moderating Role of Need for Autonomy 18 2.3 The Moderating Role of Perceived Self-efficacy 19 3 Methods 20 3.1 Participants 20 3.2 Experimental Design and Procedure 20 3.3 Operationalization 21 3.4 Measurements 22 3.5 Statistical Analysis 23 4 Results 23 4.1 Main Analysis 23 4.2 Secondary Analysis 25 5 Discussion 25 5.1 Main Findings 25 5.2 Limitations and Suggestions for Future Research 27 6 Conclusion 27 References 27 Interacting with the News Through Voice User Interfaces 30 1 Introduction 30 2 Background 32 2.1 Voice User Interface 32 2.2 Conversational Interactions with the News 33 3 Methods 33 3.1 Field Trial 33 3.2 Interviews 34 3.3 Analysis 35 4 Results 35 4.1 Overview 35 4.2 Initiating News Interactions 36 4.3 Navigating Between News Sources 37 4.4 Navigating Within a News Source 37 4.5 Accessing Specific Genres of News 38 4.6 Seeking Additional Information and Reports 39 4.7 Requesting Explanations Related to Elements in the News 39 5 Discussion 40 5.1 Interactional Guidance in News Contexts 41 5.2 Navigation Between and Within News Sources 42 5.3 In-Depth Exploration of the News Through VUIs 42 6 Conclusion 43 References 44 Voice Your Opinion! Young Voters’ Usage and Perceptions of a Text-Based, Voice-Based and Text-Voice Combined Conversational Agent Voting Advice Application (CAVAA) 46 1 Introduction 46 1.1 Modality 48 2 Method 49 2.1 Design 49 2.2 Participants 50 2.3 Materials 50 2.4 Usage Measures 53 2.5 Evaluation Measures and Political Measures 53 2.6 Procedure 54 3 Results 54 3.1 Usage Measures 54 3.2 Evaluation Measures and Political Measures 56 4 Discussion 57 5 Conclusion 58 References 59 Value Creation in Gamified Chatbot Interactions and Its Impact on Brand Engagement 62 1 Introduction 62 2 Theoretical Framework and Hypotheses Development 63 2.1 Chatbots and Gamification 63 2.2 Consumer Engagement 64 2.3 Utilitarian and Hedonic Values 65 3 Research Method 66 3.1 Study Design 66 3.2 Main Study: Measurement Development and Sample 68 4 Results 69 4.1 Demographic Information 69 4.2 PLS Analysis and Measurement Model 69 4.3 Structural Model 69 5 Discussions, Implications and Future Studies 72 References 74 Chatbots as Part of Digital Government Service Provision – A User Perspective 78 1 Introduction 78 2 Background 79 2.1 Digital Government 79 2.2 Chatbots as Part of Digital Government 79 2.3 User Perceptions of Chatbots and Usage Motivation 80 2.4 Technology Acceptance as Perspective to Understand Intention to Use 81 3 Research Question 82 4 Method 82 4.1 Participants and Recruitment 82 4.2 The Municipality Chatbot – Kommune-Kari 83 4.3 Interview Process 83 4.4 Analysis 84 5 Results 84 5.1 The Participants’ Chatbot Experience and Intention to Use 84 5.2 Performance Expectancy 85 5.3 Effort Expectancy 86 5.4 Trust 87 5.5 Hedonic Motivation 87 5.6 Habit 88 5.7 Social Influence and Facilitating Conditions 89 6 Discussion 89 6.1 Citizen Perceptions of Government Chatbots and Implications for Intention to Use 89 6.2 Implications 90 6.3 Limitations and Future Research 91 References 92 Understanding the Intention to Use Mental Health Chatbots Among LGBTQIA+ Individuals: Testing and Extending the UTAUT 95 1 Introduction 95 2 Theoretical Background 96 2.1 The UTAUT 96 2.2 Extending the Model: Willingness to Self-disclose, Perceived Loss of Privacy, Trust and Gender Identity 97 3 Method 99 3.1 Sampling 99 3.2 Pretest 99 3.3 Procedure 99 3.4 Measurements 100 3.5 Analysis 100 4 Results 101 4.1 Sample Characteristics 101 4.2 Model Fit and Hypothesis Testing 101 5 Discussion 103 5.1 Limitations and Future Research 104 Appendix 1 105 Appendix 2 107 Appendix 3 107 Appendix 4 108 Appendix 5 109 References 109 Chatbot Design and Applications 113 Enhancing Conversational Troubleshooting with Multi-modality: Design and Implementation 114 1 Introduction 114 2 Related Work 116 2.1 Dialogue Management in Conversational Agents 116 2.2 Conversational Troubleshooting 116 3 Preliminary Analysis 117 3.1 Design Principles 117 3.2 Usage Scenario 118 3.3 Problem Formalisation 119 4 System's Overview 120 4.1 Configuration 120 4.2 Architecture 122 4.3 Runtime Behaviour 123 4.4 Deployment 125 5 Conclusion 125 References 125 A Framework and Content Analysis of Social Cues in the Introductions of Customer Service Chatbots 129 1 Introduction 129 2 Theoretical Framework 130 2.1 Managing Customers’ Expectations of Customer Service Chatbots with Chatbot Introductions 130 2.2 Social Cues in Chatbot Introductions 131 3 Method 134 3.1 Sample of Chatbot Introductions 134 3.2 Codebook 134 3.3 Coding Procedure 135 3.4 Data Analysis 136 4 Results 136 4.1 Reliability of Social Cues in Chatbot Introduction 136 4.2 Usage of Social Cues in Chatbot Introductions 136 4.3 Qualitative Analysis of Social Cues in Chatbot Introductions 138 5 Conclusion and Discussion 140 References 142 An Affective Multi-modal Conversational Agent for Non Intrusive Data Collection from Patients with Brain Diseases 145 1 Introduction 145 2 Related Work 147 2.1 Conversational Agents for Health 147 2.2 Sentiment Analysis from Text and Video 147 3 Zenon Conversational Agent 148 3.1 Conversation Manager 150 3.2 Sentiment Analysis on User Generated Text 151 3.3 Sentiment Analysis on Facial Expressions 152 3.4 Sentiment Aggregation 154 3.5 The Semantic Knowledge Graph (SemKG) 154 3.6 User Interface - The Android Application 155 4 Evaluation, Feedback, and Lessons Learned 156 5 Conclusions and Next Steps 158 References 159 Interactive Journaling with AI: Probing into Words and Language as Interaction Design Materials 161 1 Introduction 161 2 Background 162 2.1 Journaling and Reflection 162 2.2 Human-AI Interaction 163 2.3 Conversation as Interaction Modality 163 3 Methodology 164 4 Designing Conversational Interactions 166 4.1 Prototyping CA’S Personality and Conversations 169 4.2 Testing Journaling Experiences with CA 169 5 Results 171 5.1 Interactive and Reflective Journaling with AI 171 5.2 Conversational and Collaborative Journaling with AI 173 6 Limitations 174 7 Concluding Discussions 174 Appendix 176 Appendix A. Examples of CA’S Key Behaviors and Prompts 176 Appendix B. The Interaction Design Wheel(s) of Fortune 177 Appendix C: Combinations of Elements of Interaction 178 References 179 Increasing the Coverage of Clarification Responses for a Cooking Assistant 182 1 Introduction 182 2 Related Work 183 2.1 Modelling Knowledge for Conversational Agents 183 2.2 Distinguishing Between Different Knowledge Sources 184 3 Increasing the Coverage of Clarification Responses 184 3.1 Commonsense Question Answering 185 3.2 Task-Specific Question Answering 187 3.3 Combining Question-Answering Methods 189 4 Analyzing the Quality of Clarification Responses in Real-world Cooking Assistant Conversations 190 4.1 Conversational Agent Architecture 190 4.2 Selecting Clarification Responses 191 4.3 Experiment Design and Data Analysis 192 4.4 Results 193 5 Discussion 194 6 Conclusion 195 A General Question-Answering: Question Types 196 B Recipe-Specific Question-Answering: Intent Types 197 C Recipe Annotation Guidelines: Ground Truth 198 References 198 Designing Context-Aware Chatbots for Product Configuration 201 1 Introduction 201 2 Theoretical Background and Related Work 203 2.1 Product Configurators 203 2.2 Conversational Interfaces for Complexity Reduction 203 2.3 Context-Awareness in Chatbots 203 3 Research Approach 204 4 Designing Context-Aware Chatbots for Product Configuration 205 4.1 Awareness of the Problem 205 4.2 Suggestion 207 4.3 Development 209 5 Evaluation 211 5.1 Experimental Design 211 5.2 Measurement Instruments 212 5.3 Evaluation Results 213 6 Discussion 214 6.1 Contribution 214 6.2 Limitations and Future Works 216 7 Conclusion 216 Appendix 217 Questions Asked in the Semi-structured Interviews 217 Final Evaluation 218 References 219 Author Index 222 This book constitutes the proceedings of the 6th International Workshop on Chatbot Research and Design, CONVERSATIONS 2022, which was held during November 2022. The 12 papers included in this volume were carefully reviewed and selected from a total of 27 submissions. They were organized in topical sections named: chatbot users and user experience; chatbot design and applications.