Recognition of coordinated adversarial behaviors from multi-source information

摘要: To successfully predict the actions of an adversary and develop effective counteractions, knowledge enemy's mission organization are needed. In this paper, we present new models algorithms to identify behaviors of adversaries based on probabilistic inference two main signatures behavior: plans (what enemy wants do) and organizations (how is organized who responsible for what). The technology allows extraction, classification, temporal tracking behavior using multi-source data, as well prescribes intelligence collection reduce ambiguity in current predictions.