Log in / Register
Home arrow Computer Science arrow Social Informatics
< Prev   CONTENTS   Next >

FlowSampler: Visual Analysis of Urban Flows in Geolocated Social Media Data

Abstract. Analysis of flows such as human movement can help spatial planners better understand territorial patterns in urban environments. In this paper, we describe FlowSampler, an interactive visual interface designed for spatial planners to gather, extract and analyse human flows in geolocated social media data. Our system adopts a graph-based approach to infer movement pathways from spatial point type data and expresses the resulting information through multiple linked multiple visualisations to support data exploration. We describe two use cases to demonstrate the functionality of our system and characterise how spatial planners utilise it to address analytical task.

Keywords: Social media analytics • Geovisualisation • Spatio-temporal analysis • Data mining • Flow maps

1 Introduction

Urban (or inter-urban) flow analysis is a particularly important subject in spatial planning that identifies territorial patterns in human movement to inform policymaking. Although many techniques have been devised to carryout such analysis [1-3], the growing volume of geolocated social media data presents spatial planners with new opportunities to formulate evidence based policies that could lead to improvements in the urban environment. A key component in analysing human movement is the notion of trajectory. A trajectory provides information about the position of a person through space and time. By analysing patterns in aggregated trajectories, spatial planners aim to identify pathways where important movement or flows occur. The insights that they gain from analysis are used to conceptualise territorial structures, such as functional urban areas [4], that ultimately determine where and how policies are enacted.

Geolocated social media data is a source of publicly accessible data that contains information, which may be extracted to study urban flows. Since such data typically contain a timestamp and can be referenced to specific user identifiers, it is reasonable to construct a social media user's trajectory based on a chronologically ordered set of geolocated data records. While this task may appear to be outwardly trivial, it can be rather challenging for spatial planners to accomplish with generic GIS software, as the data tend to be large and ill structured.

We present FlowSampler, a visual analytics system designed for spatial planners to gather and analyse urban flows in geolocated social media data. This work was motivated by the need for an interactive visual interface that would extract trajectories out of geolocated social media data and summarise them in a flow map [5]. The strength of this system is that it enables spatial planners to formulate and subsequently verify research questions by reconfiguring the interface for analysis at various spatial and temporal granularities. The interactions are carried out through a series of integrated control widgets that allows the spatial planners to directly manipulate the visualisation. We make three contributions in this paper: First, we propose a graph-based approach to construct a flow map from the trajectories extracted from a geolocated social media dataset. Then, we describe a visual analytics procedure to identify pathways with significant movement between sets of locations. Finally, we demonstrate the functionality and scalability of our system with two use cases that characterise the task this system addresses at different spatial and temporal granularities.

Found a mistake? Please highlight the word and press Shift + Enter  
< Prev   CONTENTS   Next >
Business & Finance
Computer Science
Language & Literature
Political science