基于多源数据的城市犯罪风险知识图谱研究  

Study on Knowledge Graph of Urban Crime Risk Based on Multi-source Data

作　　者：蒋耀 胡啸峰[1,2] JIANG Yao;HU Xiaofeng(School of Information and Cyber Security,People's Public Security University of China,Beijing 100038,China;Key Laboratory of Security Technology&Risk Assessment,Ministry of Public Security,Beijing 102623,China)

机构地区：[1]中国人民公安大学信息网络安全学院,北京100038 [2]安全防范技术与风险评估公安部重点实验室,北京102623

出　　处：《中国人民公安大学学报（自然科学版）》2022年第1期87-94,共8页Journal of People’s Public Security University of China(Science and Technology)

摘　　要：为全面考虑风险要素与犯罪之间的关系,基于多源数据构建城市犯罪风险知识图谱,为犯罪预防提供理论依据。首先,利用犯罪热点分析、地理加权回归(GWR)、Granger因果检验、Apriori算法等方法,挖掘犯罪风险要素及其关联关系。在此基础之上,建立知识图谱的模式层并进行知识抽取。最后,利用Neo4j图数据库进行知识存储及可视化分析。通过知识图谱,可以将不同风险要素进行关联,为犯罪风险提供微观解释,并为进一步分析不同风险要素之间的关系奠定基础,从而为犯罪风险分析、防控及预警提供决策支持。In order to fully consider the relationship between risk factors and crimes,the knowledge graph of urban crime risk based on multi-source data was constructed in this paper,which may provide theoretical basis for crime prevention.First,crime risk factors and their correlation were explored by using crime hotspot analysis,geographically weighted regression(GWR),Granger causality test and Apriori algorithm.On this basis,the pattern layer of knowledge graph was established,and then entities were extracted.Finally,the Neo4j graph database was used for knowledge storage and visual analysis.According to the knowledge graph,different risk factors could be associated to provide micro-explanation for crime risk and lay a foundation for further analysis of the relationship between risk factors,so as to provide decision support for crime risk analysis,prevention and warning.

关 键 词：城市犯罪 风险要素 知识图谱 多源数据 

分 类 号：D917[政治法律—法学]