Comunicació presentada a: the 17th Workshop on Computational Models of Natural Argument, celebrat conjuntament amb ICAIL 2017, el 16 de juliol de 2017 a Londres, Anglaterra.

In this work we propose a scheme for an Argumentative Dialogue System that allows a user to discuss a certain topic with a virtual agent using natural language. Starting from an agent vs agent case, we address the optimization of the agent strategy by formulating the problem as a Stochastic Game and show that this formalism generally allows the inclusion of additional strategical moves that are not based on the content of the argument. In a second step we propose our approach for the Natural Language Understanding of arguments by combining recent results of argumentation mining with a keyword-based mapping.