|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|156994||2017||18 صفحه PDF||سفارش دهید||6707 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Physica A: Statistical Mechanics and its Applications, Volume 482, 15 September 2017, Pages 573-584
Terrorism is a huge public hazard of the international community. Alliances of terrorist organizations may cause more serious threat to national security and world peace. Understanding alliances between global terrorist organizations will facilitate more effective anti-terrorism collaboration between governments. Based on publicly available data, this study constructed a alliance network between terrorist organizations and analyzed the alliance relationships with link prediction. We proposed a novel index based on optimal weighted fusion of six similarity indices, in which the optimal weight is calculated by genetic algorithm. Our experimental results showed that this algorithm could achieve better results on the networks than other algorithms. Using this method, we successfully digged out 21 real terrorist organizations alliance from current data. Our experiment shows that this approach used for terrorist organizations alliance mining is effective and this study is expected to benefit the form of a more powerful anti-terrorism strategy.