Research And Application Of Data Mining Technology In Vehicle Monitoring System

With the extensive application of information technology in the intelligent trainsportation, data mining technology has became one of the most important development directions of the modern intelligent transportation as the core of the vehicle monitoring, management and decision support. Data mining can be a good solution to the " information-rich but poor knowledge " issue in the road transport sector But at present in China, the application of data mining technology in vehicle monitoring system is still in its infancyThis paper described the definition, processes and classification of data mining technology, and conducted a comprehensive analysis of the various methods of cluster analysis, including partitioning methods hierarchical methods, density-based methods, grid-based methods and model-based methods. On the base of project requirements analysis, this paper proposed an overall design of the vehicle monitoring system which included the technical indicators, the overall architecture and the technology roadmap of the system.On the base of the research of data mining technology and the design of vehicle monitoring system, this paper further applied the data mining technology into the vehicle monitoring system. Firstly, this paper has fully taken the characteristics of vehicle data into account and completed the pretreatment process of vehicle data. Secondly, this paper proposed an optimization method of the Density-Based Spatial Clustering of Applications with Noise Algorithm, which was included in the density-based methods. With the advantage of clustering fast, dealing with the noise effectively and finding the spatial clustering of arbitrary shape easily, this method achieved the functions of discovering speeding-prone sections and monitoring the information of key vehicles This method also improved the accuracy of the speeding-prone sections discovery significantly. Finally, this paper showed the relevant interfaces of the vehicle monitoring system and analysed the results of discovering speeding-prone sections and monitoring the information of key vehicles, which proved to discover the sections and monitor the information effectively. The results can also provide a strong support for the decision making of regulatory agencies in the modern intelligent transportation.