Research On Application Of Improved Apriori Algorithm In Vehicle Trajectory Analysis

With the development of social economy and the acceleration of urbanization,the construction of smart cities has become an inevitable requirement for the development of urbanization.How to integrate cutting-edge science and technology such as data mining into the construction of smart cities will help improve the level of urban governance and development,and promote the modernization of urban governance systems and governance capabilities.Take the passing data as an example.When a vehicle is driving on a city road,it will capture a large amount of passing data through the bayonet point.Through big data technology such as data mining,the passing data is analyzed and processed to find the inherent Connections and laws can dig deeper rules and knowledge such as high-frequency vehicle entry and exit points,footholds,frequent driving routes,and trajectory information of fellow vehicles.If this information is used in the actual work of public security and other departments,it can provide strong technical support in diverting urban traffic,implementing vehicle control,and quickly detecting cases.Therefore,the application of association rule mining technology has important research significance for vehicle trajectory analysis..This paper deeply analyzes the current state of smart city construction,the current status of vehicle trajectory analysis and research,and the generation and development of association rule mining;introduces vehicle trajectory analysis methods,data preprocessing techniques,common data mining tools,and Apriori classic algorithm based on association rules mining.Aiming at the two major shortcomings of the algorithm that restrict the efficiency of the algorithm,such as scanning the database multiple times and generating a large number of candidate item sets,an improved Apriori algorithm based on non-empty subsets of transactions is proposed.The improved algorithm reduces the number of database scans and reduces the generation of candidate item sets.Starting from two aspects of scale,a set of non-empty subsets of a single transaction and a set of candidate items generated by the previous transaction are used to generate the candidate item set of the current transaction,calculate the support of each candidate set,and generate the largest candidate item set of the transaction transaction.Collect,delete candidate sets that do not meet the minimum support threshold to obtain a frequent item set set.Experiments verify that the improved Apriori algorithm has effectively improved the operating efficiency;finally,the improved algorithm is applied to the case analysis of suspected vehicle trajectory tracking.Taking the driving data of a hit-and-run licensed car in a certain place as the research object,the data is preprocessed with the help of data mining tools,and the car passes through the bayonet set as transaction data,distinguished by different dates,and set the minimum support and confidence To generate association rules.Through the analysis of the historical high-frequency entry and exit points and frequent driving routes of the licensed vehicles,the activity range of the suspects was sorted out,and the suspects were analyzed and judged based on the foothold.The suspects were accurately deployed and controlled and arrested.The collection of different vehicles passing through the historical time analyzed the co-authors who opened the casino with the suspect,and helped the public security organs successfully destroy the criminal gang.The research results show that the application of the improved Apriori algorithm in vehicle trajectory analysis can provide technical support for public security and other departments in implementing vehicle control and rapid case detection.