Study On Association Analysis Based On Maximum Clique And Its Application To Traffic Risk Factors

In the current information age,effective information seems to have become "invaluable".In order to find this effective information in massive amounts of data,all walks of life have invested a lot of manpower and material resources.Data mining technology came into being at this time,and association analysis is a classic technology in data mining theory.It mainly finds the implicit association rules between itemsets from a large amount of data and helps relevant personnel make appropriate decisions.The association analysis process is mainly divided into two steps: mining frequent itemsets and discovering association rules.Because mining frequent itemsets requires multiple scans of the database to calculate the support of each set,the time complexity is high,so most of the improvements to the association analysis process is based on the step of mining frequent itemsets.In this paper,the theories and methods in graph theory are used to analyze the association analysis process,specifically by introducing the method based on the maximum clique problem of neighboring points,as a sub-problem in the association analysis process,and using the methods and theories in graph theory to mine the association rules——abstract the data in the relational transaction database needed to be solved by relational analysis and the relationship between them as an undirected weighted graph.On the graph,several maximal cliques are obtained by using an algorithm based on adjacent points to solve the maximal clique.The problem of mining frequent itemsets is transformed into a problem of solving maximal cliques,and then the predecessor and successor relationships among nodes within each maximal clique are judged.The directed weighted maximum clique obtained is that the node itself frequently appears and the node for association rules with strong relationships between them,the idea of maximizing cliques can be used to filter out invalid nodes and edges in advance,and items that may generate frequent itemsets can be classified and combined,reducing the number of traversals and improving time performance.This paper analyzes the factors of traffic accidents based on the maximum clique association algorithm.First,starting from the causes of traffic accidents,topological modeling of traffic accidents.Secondly,use the maximum clique association algorithm in the constructed traffic accident topology map to find the potential association between the various factors of the traffic accident,and analyze the relationship between the various factors that caused the accident,especially the various factors and the seriousness of the accident.The relationship between the degree,get a very significant group of practical significance.Finally,obtain meaningful association rules from maximal cliques.Relevant management departments can take precautions in advance based on the results of the excavated rules and the specific conditions of current road traffic to reduce traffic accidents.