A game-theoretic approach for threats detection and intervention

Threat intervention with limited security resources is a chal-lenging problem. An optimal strategy is to effectively pre-dict attackers ’ targets (or goals) based on current avail-able information, and use such predictions to disrupt their planned attacks. In this paper, we propose a game-theoretic framework to address this challenge which encompasses the following three elements. First, we design a method to ana-lyze an attacker’s types in order to determine the most plau-sible type of an attacker. Second, we propose an approach to predict possible targets of an attack and the course of actions that the attackers may take even when the attackers ’ type-s are ambiguous. Third, a game-theoretic based strategy is developed to determine the best intervention approaches taken by defenders (security resources)