关键词: K-means算法, 基于密度, 车辆活动特征, 密度权值, 初始聚类中心, 类内类外划分指标

Abstract: Since the traditional K-means algorithm hardly predefines the number of clusters, and performs sensitively to the initial clustering centers and outliers, which may result in unstable and inaccurate results, an improved density-based K-means algorithm was proposed. Firstly, high-density trajectory data points were selected as the initial clustering centers based on the density of the trajectory data distribution and increasing the weight of the density of important points, to perform K-means clustering. Secondly the clustering results were evaluated by the Between-Within Proportion standard of clustering efficient function. Finally, the optimal number of clusters and clustering were determined according to the clustering results evaluation. Theoretical researches and experimental results showed that the improved algorithm could be better at extracting the trajectory key and keeping the key path information. The accuracy of clustering results was 28% higher compared with the traditional K-means algorithm and 17% higher compared with the Density-Based Spatial Clustering of Applications with Noise algorithm. The proposed algorithm has a better stability and a higher accuracy in the trajectory data clustering.

Key words: K-means algorithm, based on density, characteristics of vehicle activity, weight of the density, initial clustering center, Between-Within Proportion(BWP) standard