一种基于广义回归神经网络的城市入室盗窃串并案分析方法

An Analysis Method for Serial and Parallel Cases of Urban House Burglary Based on Generalized Regression Neural Network

随着城市的发展,城市人口的越来越多元化,这给城市治安带来了新的挑战,入室盗窃就是这个过程中不可调和的矛盾。我们以近几年城市室盗窃案件的案情文本数据为基础,提取入室盗窃案件的文本向量特征,基于广义回归神经网络模型,采用凝聚层次聚类算法作为回归方法,基于这一理论研究入室盗窃案件的串并方法,通过给办案民警提供入室盗窃案件的串并依据,从而提高案件的侦破率,减少群众的财产损失。

With the development of the city, the population of the city is more and more diversified, which brings new challenges to the public security of the city. Based on the text data of urban burglary cases in recent years, we extract the text vector characteristics of burglary cases. Based on the generalized regression neural network model, we use clustering algorithm as the regression method. Based on this theory, we study the serial and parallel methods of burglary cases. By providing the serial and parallel basis of burglary cases for the police, we can provide the serial and parallel basis of burglary cases high detection rate of cases and reduction of property losses of the masses.