Rami cation analysis using causal mapping

To operate in the real-world, intelligent agents constantly need to absorb new information, and to consider the ramiications of it. This raises interesting questions for knowledge representation and reasoning. Here we consider ramiication analysis in which we wish to determine both the likely outcomes from events occuring and the less likely, but very signiicant outcomes, from events occuring. To formalize ramiication analysis, we introduce the notion of causal maps for modelling \causal relationships" between events. In particular, we consider exis-tential event classes, for example presidential-election, with instances being true, false, or unknown, and directional events classes, for example inflation, with instances being increasing , decreasing or unchanging. Using causal maps, we can propagate new information to determine possible ramiications. These ramiications are also described in terms of events. Whilst causal maps ooer a lucid view on ramiications, we also want to support automated reasoning, to address problems of incompleteness, and to represent further conditions on ram-iications. To do this, we translate causal maps into default logic, and use the proof theory and automated reasoning technology of default logic. In this paper, we provide a syntax and semantics for causal mapping, and a translation into default logic, and discuss an integration of the approach with langauge engineering.