摘要
[目的/意义]利用子图模式对暴恐案件中的人员关联进行分析可以发现涉恐人员关联图中的规律,为反恐情报分析提供有效参考。[方法/过程]首先对涉恐基础数据进行预处理,保证图中各顶点的唯一性。通过计数统计出所有的频繁1-子图和频繁2-子图,然后不断迭代生成其他候选子图并筛选频繁子图,直到达到终止条件为止。[结果/结论]该方法根据反恐情报的特点进行了优化,避免了普通频繁子图挖掘中的大量图同构检测,挖掘出的频繁子图可以反映不同类别涉恐人员之间的联系规律和联系特点,发现暴恐案件线索,有效预测和打击恐怖活动。
[Purpose/significance]Frequent subgraph mining could provide rules which are found from relational graphs of terrorists in the crime of terrorism for counter-terrorism intelligence analysis.[Method/process]Before subgraph mining,the paper preprocessed the data to make sure the uniqueness of every vertex,the it counted the number of subgraph to produce frequent 1-subgraph and 2-subgraph in order.After that,repeated the process of generating other candidate subgraph and mining frequent subgraph until the terminal condition was satisfied.[Result/Conclusion]The result subgraphs could provide terrorism related clues which reflect the features of relationships among terrorists and effectively fight against terrorism.
作者
李勇男
Li Yongnan(School of Criminal Investigation and Counter Terrorism,People's Public Security University of China,Beijing,100038,China)
出处
《现代情报》
CSSCI
2019年第7期37-43,共7页
Journal of Modern Information
基金
教育部人文社会科学研究青年基金项目“基于数据挖掘的涉恐情报量化分析方法研究”(项目编号:17YJCZH098)
中国人民公安大学基本科研业务费项目“基于频繁子图挖掘的反恐情报关联图集分析方法研究”(项目编号:2019JKF331)
关键词
子图模式
反恐情报
数据挖掘
候选子图
图同构
subgraph pattern
counter terrorism intelligence
data mining
candidate subgraph
graph isomorphism
