In this talk, I will discuss the frontiers and the fruitful connections between the semantic web and argumentation theory. In particular, I will describe two approaches exploiting the connections of semantic web and argumentation techniques to address open issues in the web scenario. In particular, I will present an automatic framework we develop to support the management of argumentative discussions in wiki-like platforms. The framework is composed by (i) a natural language module, which automatically detects the arguments in natural language returning the relations among them, and (ii) an argumentation module, which provides the overall view of the argumentative discussion under the form of a directed graph highlighting the accepted arguments. I will present a number of experiments on the history of Wikipedia showing the feasibility of our approach. Moreover, I will describe how we use argumentation theory to detect possible inconsistencies arising in the answers' set returned to the user by the question-answering system over linked data called QAKIS.
22.01.2014 - 09:15