基于DBSCAN算法的恐怖主义风险评估模型——以铁路系统为例

为了对铁路系统涉恐事件进行风险管理,遏制铁路系统恐怖袭击事件的发生,提出基于DBSCAN(density-based spa-tial clustering of applications with noise)算法的铁路系统恐怖袭击风险评估方法.首先对1970—2017年发生的铁路系统恐怖袭击案件进行统计分析,然后采用DBSCAN算法对恐怖袭击发生次数、死亡人数和受伤人数3项风险评价指标进行聚类分析,最终客观计算出几类袭击方式、袭击目标和86个国家的风险.结果表明,该方法的分析过程避免了人工赋值和专家打分策略,评估结果更具客观性和真实性,适用于反恐情报工作的风险评估领域.

Terrorism Risk Assessment Model Based on DBSCAN Clustering:Taking Railway System as an Example

In order to manage the risk of terrorism incidents in railway system and prevent the occurrence of terrorist attacks in railway system, a risk assessment method was proposed for railway terrorist attack based on Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm. This method first statistically analyzed the terrorist attacks on railway systems from 1970 to 2017. On the basis of DBSCAN algorithm, it then chose the number of terrorist attacks, the number of deaths and the number of injured as the risk evaluting indexes to cluster analysis. Finally, it calculated the risk of different types of attacks, targets and 86 countries objectively. The results show that the analysis process of this method avoids manual assignment and expert scoring strategies, and the evaluation results are more objective and authentic, which is suitable for the field of counter-terrorism intelligence work risk assessment.