摘要
恐怖主义是人类的共同威胁,打击恐怖主义是每个国家应该承担的责任.对恐怖袭击事件相关数据的深入分析有助于加深人们对恐怖主义的认识,为反恐防恐提供有价值的信息支持.对恐怖袭击事件记录数据进行量化分析,利用归一加总、突变级数、熵值法、因子分析4种方法建立了恐怖袭击事件危害性综合评价体系,并用聚类分析、XGBoost模型对为宣称的恐怖事件进行归类,用Logit、SVM和随机森林三种机器学习方法对6个恐怖袭击重点地区2018年恐怖袭击案件是否增长进行预测.
Terrorism is a common threat to mankind and fighting against it is the responsibility of every country.In-depth analysis of data related to terrorist attacks can help deepen people’s understanding of terrorism and provide valuable information support for counterterrorism and terrorism prevention.Quantitative analysis in this paper,the terrorist attacks of record data,using a sum,mutation progression method,entropy method,factor analysis of four kinds of established the comprehensive evaluation system,harmful to the terrorist attacks and clustering analysis,XGBoost model for claims that classifies the terrorist event,with the Logit,SVM and random forest three machine learning methods for six terrorist attacks in key areas to predict whether the terrorist attacks of 2018 growth,at the end of the paper characteristics of terrorist incident in Russia are analyzed,The results showed that terrorist organizations/individuals were more likely to launch terrorist attacks in the run-up to the festival.
作者
李震巍
金国鹏
汪平
宋玉平
LI Zhen-wei;JIN Guo-peng;WANG Ping;SONG Yu-ping(School of Finance and Business,Shanghai Normal University,Shanghai 200234,China)
出处
《数学的实践与认识》
北大核心
2019年第16期139-146,共8页
Mathematics in Practice and Theory
基金
全国统计科学研究重点项目《非平稳高频金融数据的统计推断及实证研究》(2018LZ05)
上海师范大学第十期重点学科数量经济学(310-AC7031-19-004221)
关键词
恐怖袭击
综合评价
聚类分析
机器学习
terrorist attacks
comprehensive evaluation system
clustering analysis
machine learning
