A probabilistic approach to modelling uncertain logical arguments

摘要: Argumentation can be modelled at an abstract level using a directed graph where each node denotes argument and arc attack by one on another. Since arguments are often uncertain, it useful to quantify the uncertainty associated with argument. Recently, there have been proposals extend argumentation take this into account. This assigns probability value for that represents degree which is believed hold, then used generate distribution over full subgraphs of graph, in turn determine set admissible or extension. In order more fully understand argumentation, paper, we idea considering logic-based uncertain arguments. based models language, give constructed classical logic. We show how formalization logical relates arguments, consider number interesting classes assignments.