A Department of Homeland Security Center of Excellence led by the University of Maryland

A consortium of researchers dedicated to improving the understanding of the human causes and consequences of terrorism

Detecting Cooperation Behavior of Terrorist Organization Based on Adaboost Classifier


Detecting Cooperation Behavior of Terrorist Organization Based on Adaboost Classifier

Abstract: 

Terrorist organizations are not completely independent. However, they frequently interact with each other. Detecting the terrorist organizations cooperation behavior is quite helpful to develop effective counter-terrorism strategy. Computational intelligence techniques used to design effective data analysis method offer a way for monitoring covert terrorism activities. In this paper, we propose a cooperation behavior analysis solution based on Adaboost classifier to identify. The traditional social network method is extended by employing computational intelligence technique to generate automated analysis capacity. Variant network properties are combined by using Adaboost classification to improve the robustness of changes identification of cooperation behaviors. Finally, the method is applied to an open source dataset. The result indicates that Adaboost classifier can take advantage of multi-properties and is more robust than only using single property. It shows the potential of developing an effective method by using computational intelligence techniques in the field of public security.

Publication Information

Full Citation: 

Lin, Zihan, Duoyong Sun, Bo Li, and Sanjun Nie. 2016. "Detecting Cooperation Behavior of Terrorist Organization Based on Adaboost Classifier." 2016 9th International Symposium on Computational Intelligence and Design (ISCID) (December). http://ieeexplore.ieee.org/document/7830369/#full-text-section

Publication URL: 
Visit Website