Situation assessments using object oriented probabilistic relational models

This paper presents oriented probabilistic models (OPRMs), a new first order probabilistic language. OPRMs are a probabilistic frame-based representation language which allows the problem domain to be modeled in a structured manner in terms of objects and the relationships between them. We discuss our experience in applying object oriented probabilistic relational models to the formulation of situation assessments in the tactical military domain and also describe the potential use of the language for deriving impact assessments.