多源异构警务数据处理技术及智慧警务框架分析

Awalysis on Multi-Source Heterogeneous Policing Data Processing Technology and Intelligent Policing Framework

智慧警务融合了大数据、人工智能(Artificial Intelligent,AI)等多项技术,实现了警务资源的充分共享,打破了公安系统内部的业务壁垒,从而为警务决策提供了支持。在智慧警务背景下,如何实现海量多源异构数据的整合和利用,是当前人们关注的重点问题。因此,提出一种多源异构警务数据的处理方法,分别从数据预处理、特征提取、构建犯罪预警模型以及模型测试分析等方面出发,简要分析该处理技术。运用攻击树生成(Attack Tree Generation,ATG)算法、两路异构卷积神经网络模型(Two-channel Heterogeneous Convolutional Neural Networks,THCNN)模型和两阶段算法模型(Two Stage Algorithm,TSA),总结智慧警务的功能实现方式,以期为智慧警务的发展提供有益帮助。

Smart police fusion of big data,AI and other technologies,to achieve the full sharing of police resources,to break the business barriers inside the public security system,so as to provide necessary support for police decision-making.In the context of smart police,how to realize the integration and utilization of massive multi-source heterogeneous data is a key issue currently concerned.In this paper,a method of processing multi-source heterogeneous police data is proposed,and the processing technology is briefly analyzed from the aspects of data preprocessing and feature extraction,as well as the construction of crime early warning model,model testing and analysis.Then,using Attack Tree Generation(ATG)algorithm,Two-Channel Heterogeneous Convolutional Neural Networks(THCNN)model and Two Stage Algorithm(TSA)Model,the paper summarizes the functional realization of intelligent police,in order to provide beneficial help for the development of intelligent police.