Application Of Spatial Data Mining In Intelligent Transportation System

Intelligent Transportation System(ITS)is based on the application and development of GIS technology,which makes it possible to store a large amount of spatial data.How to make full use of these spatial data,excavate hidden traffic rules,and apply them to the management and service of urban traffic is the current research hotspot.This paper designs a BP neural network prediction algorithm combined with spatial clustering analysis,and applies it to intelligent transportation system to realize traffic flow prediction.Firstly,the traffic flow data is pre-classified,which reduces the computational complexity and has a good prediction effect.The main work is as follows.Firstly,this paper analyses the methods of spatial data mining and the characteristics of traffic flow data,and determines the feasibility of traffic flow mining.The spatial clustering k-means algorithm and BP neural network algorithm are deeply studied and analyzed,and the K-means spatial clustering algorithm is improved to make it more suitable for traffic flow data.Secondly,a BP neural network prediction algorithm combined with K-means clustering analysis is designed.Firstly,the traffic flow data are clustered and analyzed to obtain the distribution of traffic flow in the road network.According to the clustering results,a pattern database of traffic flow data is established.BP neural network algorithm is used for different pattern databases to carry out traffic flow.Through experimental analysis,the prediction accuracy and error of the algorithm are improved.Finally,an intelligent transportation system is designed and implemented,and the forecasting algorithm is applied in practice.Based on the key technology of GIS,the system uses the data interface of Baidu GIS platform to acquire data,through background calculation and analysis,relying on BIGMAP map server,realizes the functions of data processing,map matching and traffic flow prediction.The system runs well,and further verifies the intelligence of BP neural network algorithm combined with spatial clustering analysis.Feasibility and validity of traffic flow forecasting in traffic system.