Research On Crime Prediction Method Based On Intelligent Optimization Algorithm

In recent years,the scale of criminal crime has been rising,which not only brings great harm to the people's life and property security,but also destroys the harmonious and stable development environment of the city.And it is not conducive to the long-term stability of the country.Therefore,in-depth study of the temporal and spatial distribution of criminal activities and the accurate prediction of crime trends are of great significance for improving the ability of public security organs to study and judge policing,the precise fight against crime,building a beautiful,harmonious and peaceful production and living environment.This paper uses intelligent algorithms such as fuzzy logic,genetic taboo and chaotic particle swarm to study two types of crime prediction,one is the prediction of the geographical target of crime and the other is the number of cases.In addition,the crime prediction model is optimized from the two perspectives of optimization model algorithm and sample data processing.The main contents are as follows:(1)The criminal geographical target model is studied,and the basic principle of the criminal geographical target model is analyzed.Then the deficiencies of the model are pointed out.Finally,the original model is improved by using fuzzy algorithm.And using relevant examples to verify the accuracy and reliability of the new model(2)The BP neural network algorithm is used to predict the number of crimes in a long period of time,but it has disadvantages such as poor effect and large errors.From the perspective of optimization model algorithm,genetic algorithm and tabu search algorithm are used to optimize the long-term crime prediction model of BP-ANN.And the related model is verified by example.Finally,the prediction error of the model is also analyzed.(3)The short-term crime prediction based on BP neural network algorithm has the disadvantages of poor prediction accuracy and poor reliability.From the perspective of the optimization model algorithm,the particle swarm algorithm was used to optimize the short-period crime prediction model of BP-ANN.Based on this,a chaotic algorithm and particle swarm optimization algorithm to optimize the number of short-period crimes prediction model of BP-ANN are proposed.Then the example verification is carried out.Finally,two models are analyzed and compared.(4)Based on the chaotic particle swarm optimization algorithm to optimize the short period crime prediction model of BP neural network,from the point of view of sample data processing,a crime prediction model based on the wavelet threshold method of chaotic particle swarm optimization of BP neural network is proposed.The prediction results before and after wavelet threshold denoising are compared.Finally,the prediction accuracy,stability and reliability of the model after sample dataprocessing are analyzed.