CONVERSATIONS 2022 consisted of two days of sharing and discussion of chatbot research and design. The last 2-3 sessions each day were also available for online participation.
TUESDAY NOVEMBER 22 (0900-1700 CET)
- Opening. Margot van der Goot, University of Amsterdam
- Welcome to DigicomLab, Marthe Möller & Joanna Strycharz, University of Amsterdam.
- Introduction to the day. Asbjørn Følstad, SINTEF.
0945-1045. OPENING KEYNOTE – CATHERINE PELACHAUD: CONVERSING WITH SOCIALLY INTERACTIVE AGENT
In her talk, Catherine Pelachaud, Director of Research in the ISIR laboratory at Sorbonne University, will share from her research on the Socially Interactive Agent, adaptation of conversational strategies, and use of a range of multimodal communicative behaviors (read more).
1045-1115. COFFEE BREAK
1115-1215. SESSION 1 – CHATBOT USERS
Chair: Carolin Ischen (University of Amsterdam)
- Interacting with the news through voice user interfaces. Oda Elise Nordberg and Frode Guribye (full paper – 12 min)
- 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. Linwei He and Lotte Leeuwis (full paper – 12 min)
- Understanding the Intention to Use Mental Health Chatbots Among LGBTQIA+ Individuals: Testing and Extending the UTAUT. Tanja Henkel, Annemiek J. Linn and Margot J. van der Goot (full paper – 12 min)
1315-1400. PANEL DISCUSSION – EVALUATING CHATBOTS
Evaluating chatbots in research and practice – how to do useful and reliable evaluations? In the first ever CONVERSATIONS panel discussion, Michael McTear (Ulster University), Nena van As (boost.ai), Margot van der Goot (University of Amsterdam), and Elayne Ruane (UCD) share and discuss chatbot evaluation. The panel will seek to address questions such as ‘Which are key chatbot quality criteria?’, ‘What are useful evaluation approaches – and what are their limitations?’, and ‘How to do evaluation as part of chatbot development?’. The panel is chaired by Asbjørn Følstad (SINTEF).
1400-1500. SESSION 2 (ONSITE AND ONLINE) – UNDERSTANDING CONVERSATIONS
Chair: Caroline van Straten (University of Amsterdam)
- Conversation Mining for Customer Service Chatbots. Daniel Schloß and Ulrich Gnewuch (position paper – 10 min)
- An Affective Multi-Modal Conversational Agent for Non Intrusive Data Collection from Patients with Brain Diseases. Chloe Chira, Evangelos Mathioudis, Christina Michailidou, Pantelis Agathangelou, Georgia Christodoulou, Ioannis Katakis, Stratos Kontopoulos and Konstantinos Avgerinakis (full paper – 12 min)
- Enhancing Conversational Troubleshooting with Multi-modality: Design and Implementation. Giulio Antonio Abbo, Pietro Crovari, and Franca Garzotto (full paper – 12 min)
1500-1530. COFFEE BREAK
1530-1645. SESSION 3 (ONSITE AND ONLINE) – CHATBOT APPLICATIONS
Chair: Petter Bae Brandtzaeg (University of Oslo)
- Conversational Agents for Promoting Mental Health and Well-being: A Systematic Review (Work-in-Progress). Renwen Zhang, Han Li and Yi-Chieh Lee (position paper – 10 min)
- Prototyping Interactive Journaling with AI: Exploring Words and Language as Interaction Design Materials. Max Angenius and Maliheh Ghajargar (full paper – 12 min)
- Increasing the Coverage of Clarification Responses for a Cooking Assistant. Gina Stolwijk and Florian Kunneman (full paper – 12 min)
- Assessing the Natural Language Understanding Dilemma of Chatbots: Seeing or not Seeing the Forest for the Trees. Esmé Manandise and Raj Srivastava (Position paper – 10 min)
1645-1700. WRAP-UP DAY 1 AND BEST PAPER AWARD CANDIDATES (ONSITE AND ONLINE)
Drinks and dinner. After the first workshop day (November 22), we will enjoy some drinks onsite. Then, onsite participants wanting to continue the conversations, will be invited to meet up for dinner (dinner at self-cost).
Dinner location: Capital Kitchen (Weesperplein 4A, 1018 XA Amsterdam). – from 1900
WEDNESDAY NOVEMBER 23 (0845-1700 CET)
Introduction to the day. Asbjørn Følstad, SINTEF
0900-1100. GROUPWORKS – GROUP BREAKOUTS
In this breakout session, you can choose one of two engaging groupworks:
- Groupwork 1 (room REC C10.20: Attitudes, Preconceptions and Practices in Conversational AI Design. Sviatlana Höhn, Bettina Migge, Doris Dippold and Britta Schneider (read more).
- Groupwork 2 (room REC C2.06): Platforms for Chatbot Development and Research. Jan de Wit and Anouck Braggaar (read more)
1100-1215. SESSION 4 – USER EXPERIENCE
Chair: Divyaa Balaji (University of Amsterdam)
- Voice your Opinion! Young Voters’ Usage and Perceptions of a Text-based, Voice-based and Text-Voice combined Conversational Agent Voting Advice Application (CAVAA). Christine Liebrecht, Naomi Kamoen and Céline Aerts (full paper – 12 min)
- A Framework and Content analysis of Social Cues in the Introductions of Customer Service Chatbots. Charlotte van Hooijdonk, Gabriëlla Martijn and Christine Liebrecht (full paper – 12 min)
- Chatbots as Part of Digital Government Service Provision – A User Perspective. Nadia Abbas, Asbjørn Følstad and Cato Alexander Bjørkli (full paper – 12 min)
- SMILE: SMart Inclusive Living Environments supporting aging in place using eHealth solutions. Jiaxin Li and Marita Skjuve (project presentation – 10 min.)
1330-1430. SESSION 5 (ONSITE AND ONLINE) – DESIGN AND DEVELOPMENT
Chair: Rebecca Wald (University of Amsterdam)
- A virtual companion for older adults using the Rasa Conversational AI framework. Michael McTear, Sheen Varghese Marrokkie and Yaxin Bi (position paper – 10 min)
- Introducing a Framework for Designing and Evaluating Interactions with Conversational Agents. Mart Kicken, Chris van der Lee, Kim Tenfelde, Barbara Maat and Jan de Wit (position paper – 10 min)
- Designing Context-Aware Chatbots for Product Configuration. Niederer Tom, Daniel Schloß and Noemi Christensen. (full paper – 12 min)
1430-1500. COFFEE BREAK
1500-1530. INVITED TALK (ONSITE AND ONLINE) – CONTROLLABLE TEXT GENERATION FOR ALL AGES: EVALUATING A PLUG-AND-PLAY APPROACH TO AGE-ADAPTED DIALOGUE.
In this invited talk, Sandro Pezzelle, University of Amsterdam, shares experiences from research on adapting state-of-the-art conversational systems to different age groups by means of approaches to condition the generation of pre-trained language models, and whether, and to what extent, such adaptation can be perceived by automatic metrics and human speakers.
1530-1645. SESSION 6 (ONSITE AND ONLINE) – SUPPORTIVE AND ENGAGING CHATBOTS
Chair: Theo Araujo (University of Amsterdam)
- Value creation in gamified chatbot interactions and its impact on brand engagement. Susana Silva, Roberta De Cicco, Maria Levi and Maik Hammerschmidt (full paper – 12 min)
- Inclusive design of a sensitive virtual assistant to support people in vulnerable positions in their access to care. Hanneke van Heijster, Christine Liebrecht, Nadine Bol, Julia van Calis, Anneke van der Cruijsen, Marcel Dijkgraaf, Wien Feitz, Michael Muller, Emilie Tromp, Kris Bevelander and Marjolijn Antheunis (project presentation – 10 min.)
- ENRICH4ALL Project. Dimitra Anastasiou (project presentation – 10 min.)
- Hungry Enough for a Chatbot: Automation Opportunities for a Restaurant Recommender. Vineeth Ravi and Jessica Staddon (project presentation – 10 min.)