In dialogical argumentation it is often assumed that the involved partiesalways correctly identify the intended statements posited by each other,realize all of the associated relations, conform to the three acceptabilitystates (accepted, rejected, undecided), adjust their views when new and correctinformation comes in, and that a framework handling only attack relations issufficient to represent their opinions. Although it is natural to make theseassumptions as a starting point for further research, removing them or evenacknowledging that such removal should happen is more challenging for some ofthese concepts than for others. Probabilistic argumentation is one of theapproaches that can be harnessed for more accurate user modelling. Theepistemic approach allows us to represent how much a given argument is believedby a given person, offering us the possibility to express more than just threeagreement states. It is equipped with a wide range of postulates, includingthose that do not make any restrictions concerning how initial arguments shouldbe viewed, thus potentially being more adequate for handling beliefs of thepeople that have not fully disclosed their opinions in comparison to Dung'ssemantics. The constellation approach can be used to represent the views ofdifferent people concerning the structure of the framework we are dealing with,including cases in which not all relations are acknowledged or when they areseen differently than intended. Finally, bipolar argumentation frameworks canbe used to express both positive and negative relations between arguments. Inthis paper we describe the results of an experiment in which participantsjudged dialogues in terms of agreement and structure. We compare our findingswith the aforementioned assumptions as well as with the constellation andepistemic approaches to probabilistic argumentation and bipolar argumentation.
展开▼
