Entangled footprints: Understanding urban neighbourhoods by measuring distance, diversity, and direction of flows in Singapore

Abstract

Traditional approaches to human mobility analysis in Geography often rely on census or survey data that is resource-intensive to collect and often has a limited spatio-temporal scope. The advent of new technologies (e.g. geosocial media platforms) provides opportunities to overcome these limitations and, if properly leveraged, can yield more granular insights about human mobility. In this paper, we use an anonymized Twitter dataset collected in Singapore from 2012 to 2016 to investigate this potential to help understand the footprints of urban neighbourhoods from both a spatial and a relational perspective.We construct home-to-destination networks of individual users based on their inferred home locations. In aggregated form, these networks allow us to analyze three specific mobility indicators at the neighbourhood level, namely the distance, diversity, and direction of urban interactions. By mapping these three indicators of the spatial footprint of each neighbourhood, we can capture the nuances in the position of individual neighbourhoods within the larger urban network. An exploratory spatial regression reveals that socio-economic characteristics (e.g. share of rental housing) and the built environment (i.e. land use) only partially explain these three indicators and a residual analysis points to the need to explicitly include each neighbourhood's position within the transportation network in future work.

Characteristics of Jetters and Little Boxes: An Extensibility Study Using the Neighborhood Connectivity Survey

“…We next classified individuals into different groups based on their spatial connections to find common types of extensibility profiles. In prior work, the direction, magnitude, and distance of flow patterns successfully revealed typologies of places with different compositions of social groups and spatial interactions (Andris & Hardisty, 2011;Chen et al, 2021;Liu et al, 2018;Prestby et al, 2020).…”

Recognizing Emergencies and Multi-User Behavior Patterns Using Imperfect Data From Distributed Access Points. A Non-Intrusive Proof of Concept

“…• Human subsystem: Some techniques require some cooperation from humans to train machine learning models and follow specific instructions to enable the methods to work [10]- [24]. Some require humans to carry with them a wireless device or install a specific program on their mobile devices [10]- [15], [19]- [21], [23]- [33]. • Environment subsystem: Some techniques require a stationary (or even static) environment [10]- [26], [28]- [31], while some are robust to changes in the environment [27].…”