DENIED OPERATIONAL ENVIRONMENTS MULTI-ENTITY BAYESIAN NETWORKS LEARNING IN PREDICTIVE SITUATION AWARENESS Topic 3: Data, Information and Knowledge

Over the past two decades, machine learning has led to substantial changes in Data Fusion Systems throughout the world. One of the most important application areas for data fusion is situation awareness to support command and control. Situation Awareness is perception of elements in the environment, comprehension of the current situation, and projection of future status before decision making. Traditional fusion systems focus on lower levels of the JDL hierarchy, leaving higher-level fusion and situation awareness largely to unaided human judgment. This becomes untenable in today’s increasingly data-rich environments, characterized by information and cognitive overload. Higher-level fusion to support situation awareness requires semantically rich representations amenable to automated processing. Ontologies are an essential tool for representing domain semantics and expressing information about entities and relationships in the domain. Probabilistic ontologies augment standard ontologies with support for uncertainty management, which is essential for higher-level fusion to support situation awareness. PROGNOS is a prototype Predictive Situation Awareness (PSAW) System for the maritime domain. The core logic for the PROGNOS probabilistic ontologies is Multi-Entity Bayesian Networks (MEBN), which combines First-Order Logic with Bayesian Networks for representing and reasoning about uncertainty in complex, knowledge-rich domains. MEBN goes beyond standard Bayesian networks to enable reasoning about an unknown number of entities interacting with each other in various types of relationships, a key requirement for PSAW. The existing probabilistic ontology for PROGNOS was constructed manually by a domain expert. However, manual MEBN modeling is labor-intensive and insufficiently agile. To address this problem, we developed a learning algorithm for MEBN-based probabilistic ontologies. This paper presents a bridge between MEBN and the Relational Model, and a parameter and structure learning algorithm for MEBN. The methods are evaluated on a case study from PROGNOS.