Seoul's Wi-Fi hotspots: Wi-Fi access points as an indicator of urban vitality

Highlights

• Urban vitality is systemically measured in social, economic, and virtual dimensions using big data.

• Correlation among the three vitality measures is calculated based on bivariate spatial association measure, Lee's L.

• Causal relationship of urban vitality in Seoul between physical and virtual spaces is investigated by path analysis.

• The location of Wi-Fi access points is proposed as an indicator of urban vitality in virtual space.

Abstract

Advances in information and communication technology and the collection of big data at small spatiotemporal scales have opened up new opportunities to reveal spatiotemporal relationships of human dynamics and urban space. As a measure of human dynamics, however, urban vitality has been elusive both in conceptualization and measurements, and previous academic and policy attempts to grasp the concept failed to address the problems. In an effort to operationalize the concept, this paper has two goals: (1) to develop a systematic measurement of urban vitality in social, economic, and virtual dimensions and (2) to empirically investigate the relationship between physical and virtual spaces at small spatial and temporal scales. Three different indicators of urban vitality to represent the three dimensions are (1) pedestrian traffic based on cell phone activities (social vitality), (2) bank card transactions (economic vitality), and (3) the location of Wi-Fi access points (virtual vitality). To verify the correlation between the indicators, both an aspatial measure, Pearson's r coefficient, and a bivariate spatial association between the three dimensions, calculated by Lee's L measure, are used. It is also followed by path analysis to investigate the causal relationship in four path models. The empirical analysis validates the two hypotheses: (1) a spatial correlation exists between the three different urban vitality measures and (2) the spatial distribution of Wi-Fi access points could be an indicator of urban vitality. Further comparative analyses are on the condition-specific differentiation of urban vitality characteristics by time slots, age/gender groups, and regions. The virtual-physical relationship opens up possibilities for urban researchers to adopt virtual dimension as one of the ways to gauge urban vitality.

Introduction

The field of urban studies has grappled with understanding spatial and temporal relationships of human dynamics and urban space. Urban vitality has been studied as one of the components that reveal the relationship (Lynch 1981). In recent years, advances in information and communication technology (ICT) and aware technology has provided new data, methods and study subjects for researchers to investigate urban vitality. Although new indicators for systematic measurement of urban vitality have been proposed (Jin et al. 2017; Long & Huang 2017), most studies on urban vitality still have not considered virtual space and how it is intertwined with physical space. Since today's everyday life activities in urban space have both physical and virtual dimensions, it is important for urban scholars to investigate urban vitality in the two spaces and the relationship between them for a better understanding of human dynamics in cities. To fill the gap, the main focus of the paper is to measure the spatiotemporal pattern of human activities in both physical and virtual spaces. As an empirical approach, social, economic, and virtual dimensions of urban vitality are measured by big data collected in Seoul, South Korea. The study compares not only the correlation among the three vitality indicators but also examines the difference in the correlation across different regions, genders, and age groups.

The contribution of the paper is twofold: (1) developing a systematic measurement of urban vitality in virtual space and (2) empirically examining the relationship of urban vitality indicators between physical and virtual spaces at small spatial and temporal scales. To achieve the two goals together, this paper conceptualizes urban vitality as having three dimensions: social, economic, and virtual. In particular, this paper proposes the virtual dimension of urban vitality in addition to the social and economic dimensions that had been relatively well-researched. Wi-Fi (wireless local area network) access points are proposed as a proxy to represent the virtual dimension and to integrate both virtual and physical dimensions. Far from being independent of the physical space of the city, an analysis on the virtual dimension of urban vitality reveals that the virtual space is spatially embedded in the physical space. As a proxy of the virtual dimension of urban vitality, the location of Wi-Fi access points is used in the research.

This paper uses three datasets to derive a specific measure of the social, economic, and virtual dimensions of urban vitality in Seoul, South Korea: big data on pedestrian traffic, bank card transactions, and the density of Wi-Fi access points. The choice of datasets is especially pertinent in the study site of Seoul, whose virtual connectivity is one of the most vibrant in the world (Townsend 2007)¹. Using the data, this paper firstly investigates the correlation between the temporal dimension of urban vitality and Wi-Fi coverage at a block level, i.e., the smallest spatial unit of analysis of the census system. Secondly, this paper analyzes spatial, temporal, and demographic aspects of urban vitality through Wi-Fi hotspots locations. It is followed by a comparison between three city centers of high urban vitality in Seoul.

In the following sections, this paper reviews previous literature on the urban vitality and the concept's measurement in physical and virtual spaces. Upon reviewing the literature, this paper proposes a third dimension of urban vitality, the virtual dimension, in addition to social and economic dimension to expand the discourse from physical to virtual space. The following section clarifies the big data sources used in the research and describes the socio-cultural characteristics of Seoul and the selected city centers. The Methods section introduces the Lee's L (S.-I. Lee 2001), a bivariate spatial association measure, which is applied to verify the relationship among the three vitality measures for each spatial unit. The main findings of the research are presented in the Results section, and the implications and contributions of the paper are further stated in the Discussion section.