Q. Zhou

Pages: 101-108

Abstract
Mining the potential law of massive traffic flow data at intersections plays an important role in alleviating traffic pressure and optimizing traffic network. This paper introduces the clustering algorithm of data mining technology to analyze the traffic flow data, selects the main intersections of Zhenjiang City to investigate and analyze the traffic flow, and on this basis, adopts the hierarchical clustering method and K-means clustering method to cluster the traffic flow data to obtain the road spatial distribution characteristics, with the research results as follows: hierarchical clustering method can clearly divide traffic flow into three categories, the clustering result of K-means clustering method varies with the value of k, when k=3, traffic flow is divided into three categories, and k=4, traffic flow is divided into four categories; through comparative analysis, the clustering effect of hierarchical clustering method is better than K-means clustering method. The results of this study can provide a priori knowledge for predicting traffic congestion identification status and operation, and have certain practical application value.
Keywords: data mining; hierarchical clustering method; K-means clustering method; traffic flow data