Role Detection in Bicycle-Sharing Networks Using Multilayer Stochastic Block Models

TL;DR

This paper introduces novel time-dependent stochastic block models to classify bicycle-sharing stations, revealing urban mobility patterns and station roles over time in major US cities.

Contribution

The paper develops new time-dependent SBMs with degree heterogeneity for classifying nodes based on activity patterns, advancing community detection in temporal multilayer networks.

Findings

Successfully identified work, home, and other districts.

Revealed city-specific activity patterns.

Provided insights for bicycle-sharing system design.

Abstract

In urban spatial networks, there is an interdependency between neighborhood roles and the transportation methods between neighborhoods. In this paper, we classify docking stations in bicycle-sharing networks to gain insight into the human mobility patterns of three major United States cities. We propose novel time-dependent stochastic block models (SBMs), with degree-heterogeneous blocks and either mixed or discrete block membership, which classify nodes based on their time-dependent activity patterns. We apply these models to (1) detect the roles of bicycle-sharing docking stations and (2) describe the traffic within and between blocks of stations over the course of a day. Our models successfully uncover work, home, and other districts; they also reveal activity patterns in these districts that are particular to each city. Our work has direct application to the design and maintenance of bicycle-sharing systems, and it can be applied more broadly to community detection in temporal and multilayer networks with heterogeneous degrees.