基于DRF-KPCA和非线性面板模型的地区涉恐安全风险评估与影响因素研究
Regional Terrorism-Related Safety Risk Evaluation and Its Determinants Research Based on DRF-KPCA and Nonlinear Panel Model
摘要: 地区涉恐安全风险评估与影响因素研究能够研判地区未来的反恐态势，提高反恐斗争的针对性和资源的管理效率。运用空间统计方法对全球恐怖袭击事件发生的分异特征进行分析，选取恐怖袭击事件数、恐怖袭击死亡人数、恐怖袭击受伤人数和恐怖袭击财产损害程度等指标构建地区涉恐安全风险评估指标体系，提出基于密度敏感稳健模糊核主成分分析的多维面板数据的分级模型。与其他的主成分聚类算法相比较，所提算法的合理性和准确度都有提升。结合主成分方法提取出来的风险水平因子，运用最大似然估计方法建立随机效应面板顺序Logit模型，考察宏观经济和社会发展因素对地区涉恐安全风险的影响，最后对回归模型展开预测绩效评价。
Abstract: The research on regional terrorism-related safety risk evaluation and its determinants can judge the future situation of regional anti-terrorism, and improve the pertinence of anti-terrorism fight and efficiency of resource management. In this paper, the spatial statistical method is used to analyze the differentiation characteristics of global terrorist attacks. We construct the regional terrorism-related safety risk evaluation index system according to the variables such as number of terrorist attack, number of deaths, number of the injured and extent of property damage in terrorist attacks, and put forward a multi-dimensional panel data classification model based on density sensitive robust fuzzy kernel principal component analysis. Compared with other principal component clustering algorithms, the rationality and accuracy of the proposed algorithm are improved. The influence of macroeconomic and social development factors on regional terrorism-related safety risk is considered by presenting the random effect panel order Logit model based on maximum likelihood estimation method and the risk level factors obtained from principal component method. Finally, the prediction performance evaluation of the regression model is carried out.
文章引用：黄苏莉, 陆秋君. 基于DRF-KPCA和非线性面板模型的地区涉恐安全风险评估与影响因素研究[J]. 应用数学进展, 2021, 10(4): 974-988. https://doi.org/10.12677/AAM.2021.104106

参考文献

[1] 张将星. 1970-2013年全球恐怖主义活动的特征分析——兼论“9·11”事件前后时段比较[J]. 同济大学学报(社会科学版), 2016, 27(2): 48-57.
[2] Ordu, G.E.O. (2017) Trends and Patterns of Boko Haram Terrorist and Militants’ Aggression in Nigeria. Aggression and Violent Behavior, 37, 35-41. [Google Scholar] [CrossRef] 
[3] 李丽华, 韩思宁. 暴恐事件网络舆情传播机制及预防研究-英国典型案例的实证分析[J]. 情报杂志, 2019, 38(11): 102-111, 54.
[4] 刘乐. 论恐怖主义的社会解构[J]. 国际安全研究, 2019, 37(4): 30-50, 157-158.
[5] 周秋君. 恐怖主义在欧洲发展的新态势及其原因分析[J]. 社会科学, 2019, 40(2): 29-37.
[6] Oliveira, A. and Huertas, A. (2019) How Do Destinations Use Twitter to Recover Their Images after a Terrorist Attack? Journal of Destination Marketing and Management, 12, 46-54. [Google Scholar] [CrossRef] 
[7] 刘德海, 周婷婷. 基于认知差异的恐怖主义袭击误对策分析[J]. 系统工程理论与实践, 2015, 35(10): 2646-2656.
[8] 付举磊, 肖进, 孙多勇, 等. 基于社会网络的恐怖活动时空特征分析[J]. 系统工程理论与实践, 2015, 35(9): 2324-2332.
[9] 王雷, 王欣, 赵秋红. 基于和声搜索算法优化支持向量机的突发暴恐事件分级研究[J]. 管理评论, 2016, 28(8): 125-132.
[10] Estrada, M.A.R. and Koutronas, E. (2016) Terrorist Attack Assessment: Paris November 2015 and Brussels March 2016. Journal of Policy Modeling, 38, 553-571. [Google Scholar] [CrossRef] 
[11] 郭璇, 吴文辉, 肖治庭. 基于事件树和PRA的民航机场恐怖袭击风险评估模型[J]. 计算机应用研究, 2017, 34(6): 1809-1811.
[12] Buscema, M., Sacco, P.L., Massini, G., et al. (2018) Unraveling the Space Grammar of Terrorist Attacks: A TWC Approach. Technological Forecasting and Social Change, 132, 230-254. [Google Scholar] [CrossRef] 
[13] Lanouar, C. and Goaied, M. (2019) Tourism, Terrorism and Political Violence in Tunis: Evidence from Markov-Switching Models. Tourism Management, 70, 404-418. [Google Scholar] [CrossRef] 
[14] 何晶, 李本先. 基于多局域的恐怖组织网络择优增长演化模型[J]. 自动化学报, 2019, 45(11): 2137-2147.
[15] 宋艳, 滕辰妹. 基于现存设施布局的设施选址: 应对恐怖袭击[J]. 运筹与管理, 2019, 28(10): 5-12.
[16] Tao, X.M., Chang, R., Li, C.X., et al. (2019) Density-sensitive robust fuzzy kernel principal component analysis technique. Neurocomputing, 329, 210-226. [Google Scholar] [CrossRef] 
[17] 肖泽磊, 李帮义, 刘思峰. 基于多维面板数据的聚类方法探析及实证研究[J]. 数理统计与管理, 2009, 28(5): 831-838.
[18] 陈强. 高级计量经济学及Stata应用[M]. 第2版. 北京: 高等教育出版社, 2014.
[19] 李兵, 颜晓晨. 中国与“一带一路”沿线国家双边贸易的新比较优势——公共安全的视角[J]. 经济研究, 2018, 53(1): 183-197.
[20] 谢振东, 文帆, 周子雅, 等. 对恐怖袭击事件的多元统计分析[J]. 经济数学, 2019, 36(3): 89-94.
[21] 舒洪水, 李燕飙. 基于大数据视角的恐怖袭击特点与趋势分析——以GTD中的7133次恐怖袭击为样本[J]. 情报杂志, 2019, 38(11): 87-93, 198.
[22] 李国辉. 全球恐怖袭击时空演变及风险分析研究[D]: [博士学位论文]. 合肥: 中国科学技术大学, 2014.
[23] 许通. 社会分裂与恐怖主义动因探究[D]: [硕士学位论文]. 厦门: 厦门大学, 2019.
[24] Scholkopf, B., Smola, A. and Mulller, K.R. (1998) Nonlinear Component Analysis as a Kernel Eigenvalue Problem. Neural Computation, 10, 1299-1319. [Google Scholar] [CrossRef] 
[25] Heo, G., Gader, P. and Frigui, H. (2009) RKF-PCA: Robust Kernel Fuzzy PCA. Neural Networks, 22, 642-650. [Google Scholar] [CrossRef] [PubMed]
[26] Mundlak, Y. (1978) On Pooling Time Series and Cross Section Data. Econometrica, 46, 69-85. [Google Scholar] [CrossRef] 
[27] 国显达, 那日萨, 崔少泽. 基于CNN-BiLSTM的消费者网络评论情感分析[J]. 系统工程理论与实践, 2020, 40(3): 653-663.

为你推荐