An interface between natural language and abstract argumentation frameworks for real-time debate analysis

摘要

Participatory approaches are increasingly being used to plan policies, usually involving stakeholders with conflicting interests, visions and objectives. Participants are often passionate about their cause, especially in sensitive contexts such as their health or the environment for example. Whether involved in, or observing a debate, more often than not, it is hard to follow its progress; positions are unclear and arguments are unstructured often resulting in circular discussions. This is true both for participants in an ongoing debate hoping to reach a consensus as well as for those who, a posteriori, wish to understand what was discussed and how. However, at the current time, there are to our knowledge no tools to support real-time debates by allowing participants to visualize arguments and identifying opposing points of view in order to resolve conflicts. In order to fill this gap, we developed a model based on Dung's argumentation framework that we implemented in a tool called AIPA, along with a web-application to facilitate user interaction. AIPA allows one to formalize and visualize, in real-time, the arguments of the participants, to conduct inferences in order to identify acceptable arguments and to highlight conflicting ones. Furthermore, AIPA can be used a posteriori to summarize a debate in an easy to follow argument representation. The main contribution of AIPA is its ability to structure the debate by making the arguments chain traceable and transparent, in real-time and a posteriori, and this, with users with no expertise in argumentation. In this paper, we present AIPA along with two applications to illustrate its workings and benefits.