Laplacian Centricity Peak Clustering Algorithm And Its Application In Traffic District Division

Today,the standard of living of urban residents has been greatly improved,they tend to choose a convenient and comfortable way to travel,taxi is becoming a good choice for people.Because most of the taxi is equipped with GPS equipment,you can record the trajectory data,so that researchers can use these huge amounts of data to study the transport network.The traffic network as the research category of complex network science,so we can use the community structure of the subject of knowledge,division of the district city traffic network,and then analyzed the structure of the entire city's traffic network.As an important tool of data mining,clustering analysis can classify them.It is widely applied in many fields such as pattern recognition and biology.However,the algorithm also exists problems such as the initial parameter selection will affect the effect of clustering and data set type and dimension have some restrictions and so on.In this paper,we address the above questions,put forward own ideas.First,original unclassified dataset is converted into a weighted complete graph in which a node represents a data point and distance between two data points is used as weight of the edge between the corresponding two nodes.Second,local importance of each node in the network is calculated and evaluated by Laplacian centrality.The cluster center has higher Laplacian centrality than surrounding neighbor nodes and relatively large distance from nodes with higher Laplacian centralities.The new algorithm is a true parameter-free clustering method.It can classify the dataset automatically without any priori parameters.In this paper,the new algorithm was compared with 9 well-known clustering algorithms in 6 real datasets.Results show that the proposed algorithm has good clustering effect.Clustering analysis of complex networks and machine learning in the community have something in common,so we through the typical manifold learning algorithm IsoMap to reduce the dimensions of the reality of the network,so that the clustering algorithm is applied to the community,this paper combines our proposed peak clustering algorithm Laplacian based on the center division of traffic zone,and further in-depth understanding of the road network structure and traffic information of city traffic network,provide convenient traffic guidance for people's daily travel.We through the traffic of city road network,traffic congestion on the city to the region in the morning and evening peak hours and ordinary time are analyzed,in order to provide traffic planning strategy more related to the traffic management department,to facilitate their reasonable planning.