Synthesis of Argumentation Graphs by Matrix Factorization

Abstract. In the phase of evaluation of accepted arguments, one may find that not all the arguments of discussion are essential when drawing conclusions. Especially when the cardinality of the set of arguments is high, the task of identifying the most relevant arguments of the whole discussion in huge Argument Systems through the analysis of its synthesis may favor better interpretability and may allow us to extract semantics that include the the strongest arguments. We propose a new matrix interpretation of argumentation graphs and exploit a matrix decomposition technique, i.e. the Singular Value Decomposition, in order to yield a synthetized argument system with only the most prominent arguments.