Bayesian Modeling using Belief Networks of Perceived Threat levels Affected by Stratagemical Behavior Patterns

Colleen L. Phillips, John R. Crosscope, Norman D. Geddes

During a counterinsurgency campaign, there are six main sociocultural factors affecting the host nation's (HN) populace's opinion of military operational effects. These factors were used as multiple indicators of overall perceived levels of a group's threat of fear (PLT) under a variety of scenarios and conditions. Belief Networks were used to model and compute a group's PLT factor and its corresponding classification level of threat or fear. A statistical evaluation of the generated results shows that the computed PLTs do decline when a total show of US national power (use of all PMESII elements to create effective DIMEFIL effects) is exhibited. The construction and execution of the models are described and the implications are addressed.

Subjects: 3.4 Probabilistic Reasoning; 4. Cognitive Modeling

Submitted: Jun 22, 2008


This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.