铁路警务信息系统中的数据挖掘方法应用

发布时间：2018-08-12 19:23

【摘要】：随着“一带一路”战略持续推进及“高铁时代”全面到来,铁路治安管理与平安铁路建设成为广受社会关注、涉及大众切身利益的重要领域。利用海量数据分析与挖掘技术来预防及减少铁路犯罪案件发生,维护铁路治安稳定,创新防控体系,逐渐成为当前铁路警务信息化研究的重要方向。本文针对铁路警务数据进行分析处理,根据铁路犯罪的特点及规律,通过数据挖掘方法研究了铁路犯罪预警及犯罪模式识别。针对铁路犯罪预警问题,采用决策树算法来进行嫌疑人是否犯罪的判断,建立了预警分析的分类框架,基于真实铁路治安数据集进行了测试,实验结果从准确度、平均运行时间等指标验证了本文预警分析的有效性。利用关联挖掘算法识别了频繁出现的犯罪模式,有利于排查和发现犯罪嫌疑人。此外,基于警务数据集对不同条件下算法运行时间等指标进行了对比,实验结果验证了本文关联分析具有一定的效率优势。最后,本文设计并实现了铁路警务信息数据挖掘系统,包括预警分析、关联分析、统计汇总等多个模块,并通过预警及关联分析算法实验验证了挖掘系统的合理性。实验及分析结果表明:利用数据挖掘技术对铁路警务信息系统中的相关数据进行挖掘,提升了铁路犯罪预警及犯罪模式识别能力,有效遏制新的犯罪发生,为领导决策提供支持,并对警务信息系统的创新建设具有一定的指导和借鉴作用。
[Abstract]:With the continuous promotion of "Belt and Road" strategy and the arrival of "high-speed railway era", railway security management and the construction of peaceful railway have become an important field of public concern, involving the vital interests of the public. Using massive data analysis and mining technology to prevent and reduce the occurrence of railway crime, to maintain railway security and stability, to innovate the system of prevention and control, has gradually become an important direction of railway police information research. According to the characteristics and rules of railway crime, this paper studies the early warning and pattern recognition of railway crime through data mining method. Aiming at the early warning problem of railway crime, the decision tree algorithm is used to judge whether the suspect is guilty or not, and the classification framework of early warning analysis is established. Based on the real railway public security data set, the experimental results are tested, and the experimental results are based on the accuracy. The effectiveness of the early warning analysis is verified by the average running time and other indicators. The association mining algorithm is used to identify the frequently occurring crime patterns, which is beneficial to the detection and detection of criminal suspects. In addition, based on the police data set, the algorithm running time is compared under different conditions. The experimental results show that the correlation analysis in this paper has some efficiency advantages. Finally, this paper designs and implements the railway police information data mining system, including early warning analysis, association analysis, statistical summary and other modules, and through early warning and association analysis algorithm experiments to verify the rationality of the mining system. The results of experiment and analysis show that using data mining technology to mine the relevant data in railway police information system can improve the ability of early warning and pattern recognition of railway crime, and effectively curb the occurrence of new crimes. It provides support for leadership decision, and has certain guidance and reference function for the innovation construction of police information system.
【学位授予单位】：西安工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP311.13

【相似文献】