DBSCAN Clustering Algorithms for Non-Uniform Density Data and Its Application in Urban Rail Passenger Aggregation Distribution

Xiaolu Li,Peng Zhang,Guangyu Zhu

doi:10.3390/en12193722

Abstract

With the emergence of all kinds of location services applications, massive location data are collected in real time. A hierarchical fast density clustering algorithm, DBSCAN(density based spatial clustering of applications with noise) algorithm based on Gauss mixture model, is proposed to detect clusters and noises of arbitrary shape in location data. First, the gaussian mixture model is used to fit the probability distribution of the dataset to determine different density levels; then, based on the DBSCAN algorithm, the subdatasets with different density levels are locally clustered, and at the same time, the appropriate seeds are selected to complete the cluster expansion; finally, the subdatasets clustering results are merged. The method validates the clustering effect of the proposed algorithm in terms of clustering accuracy, different noise intensity and time efficiency on the test data of public data sets. The experimental results show that the clustering effect of the proposed algorithm is better than traditional DBSCAN. In addition, the passenger flow data of the night peak period of the actual site is used to identify the uneven distribution of passengers in the station. The result of passenger cluster identification is beneficial to the optimization of service facilities, passenger organization and guidance, abnormal passenger flow evacuation.

Highlights

With the widespread popularity of GPS, smart phones and other location-aware devices, various kinds of location services and mobile social network applications continue to emerge, accumulating a large number of geographic location data, such as user check-in data, vehicle GPS trajectory, micro-blog, Twitter and other massive location data [1]
The Gaussian distributed probability density function is used to implement the dataset delamination, which eliminates the effect of uneven density distribution on the clustering effect
Through accurate clustering analysis of passenger location data in the station, can the aggregation of conventional areas be clearly presented, and real-time passenger hot spots can be found in the station, so that the operators can organize and guide the behavior of passenger groups

Summary

Introduction

With the widespread popularity of GPS, smart phones and other location-aware devices, various kinds of location services and mobile social network applications continue to emerge, accumulating a large number of geographic location data, such as user check-in data, vehicle GPS trajectory, micro-blog, Twitter and other massive location data [1]. Applications based on these location data have great application value in areas such as traffic management and control [2], recommendation systems [3], advertising [4], public facility location and assessment, abnormal population identification and evacuation [5]. Literature [13] discover relevant interest point patterns by mining users’ GPS trajectories, and find popular tourist routes; literature [14] mine personal historical location data to realize user’s friend recommendation and scenic spot recommendation

Objectives

Results

Conclusion