摘要
通过分析和挖掘出恐怖组织内在的关联特征,得出恐怖主义袭击事件组织者的时空活动规律,为针对特定组织者的全球反恐战略部署提供理论依据和指导。利用20年的全球恐怖主义数据库(GTD)作为数据源,以全球恐怖组织为主体研究对象,通过改进的多值属性Apriori算法对提取出的恐怖组织时间、空间及其相关特征进行关联分析,并以粒子群算法(PSO)优化Apriori算法的支持度和置信度两个重要参数。研究结果表明,改进算法规则提取时间有所缩短,冗余规则数量大大减少,特定恐怖组织在时空分布上具有很强的内在关联特征。由此得出结论,通过对多值属性Apriori算法的剪枝步和连接步设定规则限制能够提高关联算法的运行效率并提取出更加有效的规则。同时,经过粒子群算法的优化能够避免人为主观意识对算法结果产生的影响,从而验证了改进算法的有效性和准确性,挖掘出恐怖组织的基本时空活动规律。
By analyzing and excavating the intrinsic characteristics of terrorist organizations, the law of time and space activities of terrorist attack organizers is obtained, which provides theoretical basis and guidance for the deployment of global counter-terrorism strategies for specific organizers. Using the 20-year Global Terrorism Database(GTD) as the data source and the global terrorist organization as the main research object, the improved multi-valued attribute Apriori algorithm is used to analyze the extracted time, space and related features of the terrorist organization. Two important parameters of support and confidence of Apriori algorithm are optimized by particle swarm optimization(PSO). The research results show that the extraction time of improved algorithm rules is shortened, the number of redundant rules is greatly reduced, and the specific terrorist organization has strong internal correlation characteristics in time and space distribution. It is concluded that by setting the rule limits on the pruning step and the connection step of the multi-value attribute Apriori algorithm, the running efficiency of the association algorithm can be improved and more effective rules can be extracted. At the same time, the optimization of particle swarm optimization algorithm can avoid the influence of human subjective consciousness on the algorithm results, thus verifying the effectiveness and accuracy of the improved algorithm and mining the basic temporal and spatial activities of terrorist organizations.
作者
曾本冲
万旺根
Zeng Benchong;Wan Wanggen(School of Communication and Information Engineering,Shanghai University,Shanghai 200072,China;Institute of Smart City,Shanghai University,Shanghai 200072,China)
出处
《电子测量技术》
2020年第1期46-51,共6页
Electronic Measurement Technology
基金
上海市科委国际合作项目(18510760300)资助。
