Recent years have seen the rise of conversational agents (CAs) in everyday life: chatbots that use written communications above all, but also vocal digital assistants. It is safe to say that conversation is becoming a key mode of human-computer interaction. However, despite much recent success in natural language processing and dialogue research, the communication between a human and a machine is still in its infancy. In this context, dialogue personalization could be a key to narrow part of the gap, making sense of users’ features (e.g., preferences, expertise, communication style, emotions, personality) when engaged in a conversation with a machine. Learning user features directly from the dialogue with the chatbot, in order to adapt its response (e.g. the complexity of the explanations) can be an opportunity to improve the interaction with the user, both in commercial chatbots and voicebots offered by companies to their customers and by conversational, recommender systems.
The PICA workshop focuses on both long-term engaging spoken dialogue systems and text-based chatbots, as well as conversational recommender systems. Papers can be about different approaches to this: (pilot) evaluations, design guidelines, personalization, natural language processing, protection of privacy and (health) data, (cognitive) architectures and frameworks, implementations, context analyses. We are also very interested in studies on the effectiveness of behaviour change support systems and changing health related behaviour (such as quit smoking, lose weight, etc.).
The main goal of the workshop is to stimulate discussion around problems, challenges, possible solutions and research directions regarding the exploitation of NLP and ML techniques to learn user features and use them to personalize the dialogue in the next generation of intelligent conversational agents.
Submissions are encouraged but not limited to the following topics:Applications for personalized dialogues
Submissions to the workshop are in the form of a long or short paper and submitted as a PDF file. We accept the following submissions:
Papers have to be submitted via the EasyChair conference system (select the ‘Workshop-PICA’ track within EasyChair), must be written in English and contain original material that has not been published or is currently undergoing review elsewhere. Papers are peer-reviewed by a committee of experts in the field. Accepted papers will be listed on DBLP and published via the ACM Digital Library.
Please note that for a contribution to be included in the workshop program, and the corresponding paper to appear in the Proceedings, at least one of its authors must register and attend the workshop to present their contribution.
The paper format should follow the new ACM single-column format, please use one of the following templates:
Your submission should be close to camera-ready and thoroughly copy-edited due to the short selection cycle. Furthermore, due to the very rapid selection process we cannot offer any extensions to the deadline. Submissions are single blind and should therefore include all author names, affiliations and contact information and references to external data, software or videos, if applicable.