Stackelberg security game (SSG) has been widely used in counter-terrorism, but SSG is not suitable for modeling opportunistic crime because the criminals in opportunistic crime focus on real-time information. Hence, the opportunistic security game (OSG) model is proposed and applied in crime diffusion in recent years. However, previous OSG models do not consider that a criminal can cooperate with other criminals and this situation is very common in real life. Criminals can agree to attack the selected multiple targets simultaneously and share the utility. The police may be unable to decide which target to protect because multiple targets are attacked at the same time, so criminals can gain more utility through cooperation and interfere with police decisions. To overcome this limitation of previous OSG model, this paper makes the following contributions. Firstly, we propose a new security game framework COSG (Cooperative Opportunistic Security Game) which can capture bounded rationality of the adversaries in the cooperative opportunistic crime. Secondly, we use a compact form to solve the problem of crime diffusion in the cooperative opportunistic crime. Finally, extensive experiments to demonstrate the scalability and feasibility of our proposed approach.
CITATION STYLE
Zhao, Y., Li, M., & Guo, C. (2020). Developing Patrol Strategies for the Cooperative Opportunistic Criminals. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 11944 LNCS, pp. 454–467). Springer. https://doi.org/10.1007/978-3-030-38991-8_30
Mendeley helps you to discover research relevant for your work.