Decision making processes, be it at the societal or at the personal level, eventually come to a point where one side will challenge the other with a why-question, which is a prompt to justify one's stance. Thus, technologies for argument mining and argumentation processing are maturing at a rapid pace, giving rise for the first time to argument retrieval. We invite you to participate in the first lab on Argument Retrieval at CLEF 2020 (paper) featuring two tasks.
Project Debater is the first AI system that can meaningfully debate humans on complex topics. As an IBM grand challenge, the system can help people build persuasive arguments and make well-informed decisions. In this talk, first Yufang will briefly tell the story of Project Debater from an industry researcher's perspective. Then she will focus on its underlying technology on argument retrieval.
This talk revolves around debate technology as a way to track and analyze the process of public decision-making, the exchange of arguments, and the stance of individual participants to the points raised in a debate. Annette will present some of the insights for implementing such a real-time system in the general public and give an overview of the key linguistic properties in natural, spontaneous communication and the challenges and opportunities related to it for mining dialogue structures. A particular focus will be on conventional implicatures that is crucial in this respect: They are highly frequent in natural communication and offer strong signals of propositional boundaries and inter-propositional relationships - the two key aspects of argumentation mining. Annette will also showcase how an interdisciplinary collaboration with Visual Analytics allows us to explore the dynamics of this type of communication on a large scale and conclude with an outlook on the potential for public debate technology in other scenarios (e.g., public service encounters, education platforms) and some of the ways with which NLP can empower the general public.