为揭示恐怖组织的袭击行为模式,对袭击数量超过500件的恐怖组织袭击行为进行了统计实证。在袭击时间间隔服从幂律分布的假设下,采用最大似然估计法对其幂律指数进行估计,并采用Kolmogorov-Smirnov检验法进行真伪性检验。结果表明:各组织发动袭击的时间间隔均服从幂指数为1.8~3.1的幂律分布;恐怖袭击的阵发性与恐怖组织创立的早晚、生命周期的长短无关;组织平均活跃性与幂律指数并不具有正相关关系,组织规模的大小是影响其幂律指数的重要因素。
To reveal the attack pattern, terrorism groups which attacked more than 500 events were empirically studied. Under the hypothesis that their interevent time follow power-law distribution, the maximum-likelihood estimation was applied to estimate the power-law exponent, and then the Kolmogorov-Smirnov（KS) statistic was used to test the hypothesis. The results show that the interevent time distributions of all the groups are power-law; The burst of terrorism attacks has no relationship with the foundation time and the lifetime; There is no positive correlation between average activity and power-law exponent and the group size is the main factor to affect it.
[1] Barabási A L. The origin of bursts and heavy tails in human dynamics[J]. Nature, 2005, 435:207-211.
[2] Oliveira J G, Barabasi A L. Human dynamics :Darwin and Einstein correspondence pat terns[J]. Nature, 2005, 437:1251-1253.
[3] 樊超, 郭进利, 韩筱璞, 等. 人类行为动力学研究综述[J]. 复杂系统与复杂性科学, 2011, 8（2):1-17.
Fan Chao, Guo Jinli, Han Xiaopu, et al. A review of research on human dynamics[J]. Comples Systems and Complexity Science, 2011, 8（2):1-17.
[4] 闫小勇. 人类个体出行行为的统计实证[J]. 电子科技大学学报, 2011, 40（2):168-173.
Yan Xiaoyong. Empirical statistics on individual human travel behavior[J]. Journal of University of Electronic Science and Technology of China, 2011, 40（2):168-173.
[5] 朱军芳. 网络上集体行为的动力学研究[D].合肥:中国科学技术大学,2010.
Zhu Junfang. Dynamics research on network behaviours[D]. Hefei:University of Science and Technology of China, 2010.
[6] Clauset A, Young M. Scale invariance in global terrorism[DB/OL].[2013-01-20].http://arxiv.org/pdf/physics/0502014.pdf.
[7] Spagut M, Johnson N F, Restrepo J A, et al. Universal patterns underlying ongoing wars and terrorism[DB/OL].[2013-01-20].http://arXiv:physics/0605035v1.
[8] Zhu J F, Han X P,Wang B H. Statistical property and model for the inter-event time of terrorism attacks[J]. Chin Phys Lett, 2010, 27（6):068902.
[9] Global terrorism database[DB/OL].[2011-10-25].http://www.start.umd.edu/gtd.
[10] Asal V, Rethemeyer R K. The nature of the beast:organizational structures and the lethality of terrorist attacks[J]. Journal of Politics, 2008, 70（2):437-449.
[11] 龚小庆,王展. 关于Zipf律的一点注记[J]. 复杂系统与复杂性科学, 2008, 5（3):73-78.
Gong Xiaoqing, Wang Zhan. A note on the Zipf's law[J]. Comples Systems and Complexity Science, 2008, 5（3):73-78.
[12] Newman M E J. Power laws, Pareto distributions and Zipf’s law[J]. Contemporary Physics, 2005, 46（5):323-351.
[13] Clauset A, Shalizi C R, Newman M E J. Power-law distribution in empirical data[J]. SIAM Review, 2009, 51（4):661-703.
[14] 姚灿中,杨建梅. 幂律拟合的进展及其在产业网络中的应用[J]. 管理学报, 2008, 5（3):371-375.
Yao Canzhong, Yang Jianmei. Power-law fitting problems and application to several industrial networks[J]. Chinese Journal of Management, 2008, 5（3):371-375.
[15] 樊超, 郭进利, 纪雅莉, 等. 基于图书借阅的人类行为标度律分析[J]. 图书馆情报工作, 2010, 54（15):35-39.
Fan Chao, Guo Jinli, Ji Yali, et al. Analysis of human behavior scaling law based on library loans[J]. Liberay and Information Service, 2010, 54（15):35-39.
[16] Zhou T, Kiet H A T, Kim B J, et al. Role of activity in human dynamics[J]. Europhysics Letters, 2008, 82（2):28002.
