刑事案件关联分析与防控警务模式研究

发布时间：2018-05-18 14:08

本文选题：刑事案件 + 数据挖掘　；参考：《大连理工大学》2010年博士论文

【摘要】： 伴随着社会政治、经济和科技的高速发展,犯罪态势也随之发生变化。为了更好地保障人民群众的生产生活安全,构建和谐社会,公安机关和相关学者需要研究新形势下刑事案件的发案规律,根据发案规律和特点调整布置警力资源,最大效率地提高刑事案件的防控能力,做到警力资源的有的放矢。为了探求刑事案件的发案规律,从而科学高效地配置警力,实现事前预防和精确打击,本文做了如下的工作： (1)在界定公安管辖的刑事案件的数据内容及确定本文研究的刑事案件数据的分类原则基础上,对刑事案件的数据内容进行了分类整理,并分析了刑事案件的数据特征,总结归纳了刑事案件数据应用中存在的问题。 (2)为了描述刑事案件各方面的指标与要素,支持多源、异构数据的整合以及数据的深层挖掘与分析,构建了公安刑事案件本体；并基于所构建的公安刑事案件本体,采用访谈法,将领导与基层干警作为访谈对象,相应地分两个阶段进行了访谈,从而进一步明确了公安刑事案件防控中的问题,确定了重点分析的指标范围；之后,为了有效地支持公安刑事案件相关数据的深层次分析与挖掘,从架构角度,提出了基于本体的数据整合方法。 (3)针对公安刑事案件本体中各类间的attribute-of关系,提出了面向刑事案件的多维关联分析模型,并给出了相应的算例；针对刑事案件本体中各类间的kind-of关系,提出了面向刑事案件的多层关联分析模型,并给出了相应的算例；针对关联挖掘结果中所存在的大量冗余或现实指导意义不强的关联规则的问题,提出了一种基于语义规则的刑事案件关联规则过滤方法。基于所明确的分析指标,综合利用上述模型与方法,进行了大连市刑事案件应用实例的关联规则挖掘与结果分析。 (4)在分析原有案件防控警务模式的基础上,基于所构建的刑事案件本体和所确定的分析指标,利用统计分析和关联分析等数据挖掘手段与方法,提出了刑事案件防控警务模式的构建模型。并将该模型应用到大连市的案件防控实践中,详尽设计了围绕高发案时间段的“实效巡逻”警力投放模式、基于高发案地区和部位的“重点防控”模式、基于高发案人群的“精确管理”模式和基于被盗、抢、骗涉案物品的“动态防控模式”四种新型的案件防控警务模式。这些新型防控警务模式在大连市的应用结果表明,它可以有效地遏制案发增长趋势并提高相关类型案件的破案率。
[Abstract]:With the rapid development of social politics, economy and science and technology, the crime situation has also changed. In order to ensure the safety of the people's production and life and build a harmonious society, public security organs and relevant scholars need to study the law of criminal cases in the new situation and adjust the police resources according to the law and characteristics of the crime. To improve the ability of prevention and control of criminal cases effectively and achieve a definite target for police resources, in order to find out the law of criminal cases, so as to scientifically and efficiently deploy police force and realize pre incident prevention and accurate strike, the following work has been done.
(1) on the basis of defining the data content of criminal cases of public security jurisdiction and determining the classification principle of the data of criminal cases studied in this paper, the data content of criminal cases is classified and classified, and the data characteristics of criminal cases are analyzed, and the problems in the application of criminal cases are summarized and summarized.
(2) in order to describe the indicators and elements of all aspects of criminal cases, support multi source, heterogeneous data integration and the deep mining and analysis of the data, the public security criminal case ontology is constructed. Based on the construction of the public security criminal case ontology, the interview method is adopted to take the leadership and the basic level police as the interviewees, and then be divided into two stages accordingly. In order to effectively support the deep analysis and mining of the relevant data of public security criminal cases, the method of data integration based on the ontology is put forward, in order to effectively support the deep analysis and mining of the relevant data of public security criminal cases.
(3) in view of the attribute-of relations among the public security criminal cases, a multi-dimensional correlation analysis model for criminal cases is proposed and the corresponding calculation examples are given. In view of the kind-of relations among the various criminal cases, a multi-layer relational analysis model for criminal cases is proposed, and the corresponding calculation examples are given. This paper presents a filtering method of association rules for criminal cases based on semantic rules, which is based on semantic rules, and uses the above models and methods to carry out the mining of association rules for the application examples of criminal cases in Dalian. The results were analyzed.
(4) on the basis of the analysis of the original case prevention and control policing mode, based on the construction of the criminal case ontology and the determined analysis index, using the data mining methods and methods, such as statistical analysis and correlation analysis, the construction model of the criminal case prevention and control policing model is put forward, and the model is applied to the case prevention and control practice in Dalian. Based on the "key prevention and control" mode of the high case area and part, the "accurate management" model of the high case population and the "dynamic prevention and control mode" based on the stolen, robbed, and deceived items are four new prevention and control policing mode based on the "key prevention and control" mode of the high case area and part of the high case. The application of police control mode in Dalian shows that it can effectively curb the growth trend of cases and improve the detection rate of related cases.
【学位授予单位】：大连理工大学
【学位级别】：博士
【学位授予年份】：2010
【分类号】：D917

【引证文献】