Time-Series Representation and Clustering Approaches for Sharing Bike Usage Mining

Duo Li,Yifei Zhao,Yan Li

doi:10.1109/access.2019.2958378

Duo Li, Yifei Zhao + Show 1 more

Open Access

https://doi.org/10.1109/access.2019.2958378

Copy DOI

Abstract

Massive bike-sharing systems (BSS) usage and performance data have been collected for years over various locations. Nevertheless, researchers encountered several challenges while dealing with massive BSS data. The challenges that could be enhanced in the previous studies are 1) reducing high dimensionality and noise of BSS time series data and 2) extracting informative usage patterns out of massive BSS data. This paper extracts patterns and reduce data dimensions of BSS usage by exploring time series representation and clustering of BSS usage data. A reduced dimension allows us to efficiently approximate the BSS usage with reasonable accuracy, which can be further used for bike usage clustering, classification and prediction. We employ a non-data adaptive representation technique -Discrete Wavelet Transform (DWT) to reduce dimensionality and filter out random errors of the raw time series. Time series are clustered using k-means based on similarities measured by Dynamic Time Warping (DTW) and prototypes computed using DTW barycenter averaging (DBA). The proposed approaches are applied on a 3-month bike usage dataset acquired on the BSS of Chicago. The analysis results show that DWT can effectively reduce dimensionality, filter out random errors and reveal the main characteristics of the raw time series. The clustering approach offers the ability to differentiate and discover bike usage patterns across different stations.

Highlights

As an affordable, convenient, and sustainable travel option with various benefits, BSSs have received increasing attention in the past decade
We reduced the dimensionality of the raw dataset by 4 times using Discrete Wavelet Transform (DWT), which means that bike usage of each day was presented by 6 data points in the transformed dataset instead of 24 data points in the raw dataset
SUMMARY AND CONCLUSION To achieve a better understanding of BSS usage and performance, various attempts have been made to investigate the data collected from BSS

Summary

INTRODUCTION

Convenient, and sustainable travel option with various benefits, BSSs have received increasing attention in the past decade. Bike-sharing differs from conventional public transit (e.g., subways and buses) The former provides transportation based on demand with a decentralized structure, while the latter is operated following a regular schedule and pre-determined routes [1]. D. Li et al.: Time-Series Representation and Clustering Approaches for Sharing Bike Usage Mining incentives for distribution of bikes and obtaining a better understanding of BSS. The contributions of this work include: 1) to the best of our knowledge, this is the first work on dimension reduction of count series data Such a data representation approach leads to significant reduction of noise and computational cost during the BSS analysis. An integrated time series clustering method consisting of DTW, DBA and k-means is proposed to cluster stations over the network

RELATED WORK

CASE STUDY

SUMMARY AND CONCLUSION