移动簇在反恐情报轨迹数据聚集特征分析中的应用研究  

Research on the Application of Moving Cluster in the Clustering Feature Analysis of Counter-Terrorism Intelligence Trajectory Data

作　　者：李勇男[1] Li Yongnan(School of National Security,People's Public Security University of China,Beijing 100038)

机构地区：[1]中国人民公安大学国家安全学院,北京100038

出　　处：《情报杂志》2023年第4期57-62,共6页Journal of Intelligence

基　　金：北京市社会科学基金项目“大数据背景下北京反恐风险特征识别及应急防范机制研究”(编号:21GLC064);中国人民公安大学基本科研业务费重大项目“基于群智协同的大数据驱动的‘情景-应’’型反恐决策机制研究”(编号:2021JKF106)的研究成果之一。

摘　　要：[研究目的]对恐怖分子的时空轨迹数据进行聚集模式挖掘,有助于了解涉恐人员的群体行为模式,进而预测并提前防范一些涉恐的群体活动。[研究方法]通过对比反恐情报数据特征和多种聚集模式的特点,选择移动簇作为反恐情报中的聚集模式类型,修改移动簇挖掘经典方法中的数据采样、数据预处理、移动对象邻近度度量方法、时间点聚类方法以及移动簇的筛选条件等环节,找出适用于反恐工作需求的移动簇,再由有经验的情报分析人员进一步研判并提出处置对策。[研究结论]提出的方法通过与其他时空类型数据分析方法相结合,可以挖掘反恐情报时空轨迹数据分布规律,为反恐预警和反恐决策提供科学客观的参考。[Research purpose] Mining gathering pattern of spatiotemporal trajectory contributes to find the behavior patterns of terrorists,and then monitor and anticipate group events.[Research method] By comparing the counter-terrorism intelligence data characteristics and the characteristics of various types of gathering pattern,this paper selects moving cluster as the adaptable type.Several steps in the classic mining process need to be modified to match the requirement of counter-terrorism work.These steps include data sampling,data preprocessing,the measurement way of moving objects proximity,and the filter condition.After the mining process,experienced intelligence analysts should further study and put forward countermeasures based on the clues expressed as moving cluster.[Research conclusion] By combining with other spatiotemporal data analysis methods,this method could find the distribution law of spatiotemporal trajectory data related to counter-terrorism intelligence,and provide scientific and objective references for counter-terrorism early warning and counter-terrorism decision-making.

关 键 词：反恐情报 反恐预警 情报分析 时空轨迹 聚集模式 移动簇 

分 类 号：G359[文化科学—情报学] D631[政治法律—政治学]