Probabilistic situations for reasoning

One of the most substantial advantages that human analysts have over machine algorithms is the ability to seamlessly integrate sensed data into a situation-based internal narrative. Replicating an analogous internal representation algorithmically has proved to be a challenging problem that is the focus of much current research. For a machine to more accurately make complex decisions over a stable, consistent and useful representation, situations must be inferred from prior experience and corroborated by incoming data. We believe that a common mathematical framework for situations that addresses varying levels of complexity and uncertainty is essential to meeting this goal. In this paper, we present work in progress on developing the mathematics for probabilistic situations.