Analysis of Spatial Distribution of China's station-free bike-sharing by Clustering Algorithms

Abstract

Recently, the unbalanced spatial distribution and low utilization rate of the sharing bicycles have caused a serious impact on urban traffic. This study aims to use four clustering algorithms, including k-means clustering algorithm (KM), ant colony clustering algorithm (ACO), fuzzy c-means clustering algorithm (FCM) and mean shift clustering algorithm (MS), to analyze the data of five different density gradients. By analyzing the relationship between the density of bicycle distribution and geographical location, we can get the characteristics of the spatial distribution of shared traffic resources. In this paper, Experimental results have compared the performance of the four clustering algorithms and obtained the best algorithm for the spatial data of China's station-free bike-sharing system.