Research And Application Of Data Mining In Highway Prevent Evasion

With the implementation of the toll highway network in Guangdong Province, inthe operational management of the highway, between the road, toll collectors,supervisors, drivers according to their own interests to the risk of toll fraud evasion,affecting the recovery of highway tolls. Meanwhile, the daily toll stations producelarge amounts of charge data, and how this part of the data analysis and processing,which exists to explore the risk of evasion, highway operations management personnelare facing problems.Based on SQL Server2008provides data mining functionality to the datawarehouse, online analytical processing and data mining technology, to build tolltraffic forecasting system, implemented by the amount of vehicles and toll forecast andthe actual situation Comparative analysis of evasion behavior may exist among mining.Articles based on the characteristics of highway toll data, the overall structural designof freeway traffic forecasting system, and analysis of the current data collection systemto achieve system status data preprocessing functions. Then in the field of data analysis,the introduction of fees OLAP technology, the establishment of the cube model, so thateach toll station traffic each model can be viewed from different perspectives,compare.The paper uses the Microsoft Association Rules algorithm to mine the rulesfrom different vehicle and different time bucket,achieve association analysis oftoll-gate entrances and exits. Then the paper discusses how to properly combine theART-XP algorithm to establish traffic forecasting model meeting highwaycharacteristics,with a linear regression algorithm to predict the toll for a road,usingEuclidean distance Outlier data mining abnormal charge data. TO analyze the results ofrelevant models and algorithms with actual production data, we can summed upevasion phenomenon may exist in the daily operation of toll motorway,then takemanagement audit means perfect. Introduction of data mining analysis techniquescaneffectively reduce the loss of tolls to restore some losses for highway enterprises andthe state.