Using enthymemes in an inquiry dialogue system

• Authors: Elizabeth Black;Anthony Hunter
• Venue: Proceedings of the 7th international joint conference on Autonomous agents and multiagent systems - Volume 1
• Year: 2008

A common assumption for logic-based argumentation is that an argument is a pair (Φ, α) where Φ is a minimal subset of the knowledgebase such that Φ is consistent and Φ entails the claim α. However, real arguments (i.e. arguments presented by humans) usually do not have enough explicitly presented premises for the entailment of the claim (i.e. they are enthymemes). This is because there is some common knowledge that can be assumed by a proponent of an argument and the recipient of it. This allows the proponent of an argument to encode an argument into a real argument by ignoring the common knowledge, and it allows a recipient of a real argument to decode it into the intended argument by drawing on the common knowledge. If both the proponent and recipient use the same common knowledge, then this process is straightforward. Unfortunately, this is not always the case, and this raises interesting issues for dialogue systems in which the recipient has to cope with the disparities between the different views on what constitutes common knowledge. Here we investigate the use of enthymemes in inquiry dialogues. For this, we propose a generative inquiry dialogue system and show how, in this dialogue system, enthymemes can be managed by the agents involved, and how common knowledge can evolve through dialogue.