Towards a Common Ontology for Improved Traceability of Human Behavior Models

摘要

As human behavior models (HBMs) increase in complexity to meet escalating fidelity requirements, it has become more difficult to predict and assess the dynamic behavior of these models with any degree of depth or certainty. Consequently, a need exists for effective analysis and visualization tools that provide model traceability to support the model developer, analyst, and end-user in gaining detailed understanding of the underlying complex behaviors of HBMs. A common ontology for HBMs will allow different HBMs to be analyzed and compared using the same terminology, even when their underlying implementations may differ substantially. We are currently developing a set of model traceability tools that separate the tracing tools from the underlying HBM architecture, by employing the common HBM ontology.