Portraying the spatial dynamics of urban vibrancy using multisource urban big data. (March 2020)

• Record Type: Journal Article
• Authors: Tu, Wei; Zhu, Tingting; Xia, Jizhe; Zhou, Yulun; Lai, Yani; Jiang, Jincheng; Li, Qingquan
• Is Part Of: Computers, environment and urban systems. Volume 80(2020)
• Publication Date: 2020-03
• Subjects: Urban vibrancy -- Geographically weighted regression -- mobile phone data -- Social media -- Points-of-interest -- Big data; City planning -- Data processing -- Periodicals; Regional planning -- Data processing -- Periodicals; 303.4834
• DOI: 10.1016/j.compenvurbsys.2019.101428
• Languages: English
• ISSNs: 0198-9715

Abstract

Highlights: Three alternative urban datasets are presented to portray urban vibrancy. A comparative empirical study is conducted to examine multisource urban big data. The results demonstrate the commonalities and differences in urban vibrancy proxies. Abstract: Understanding urban vibrancy aids policy-making to foster urban space and therefore has long been a goal of urban studies. Recently, the emerging urban big data and urban analytic methods have enabled us to portray citywide vibrancy. From the social sensing perspective, this study presents a comprehensive and comparative framework to cross-validate urban vibrancy and uncover associated spatial effects. Spatial patterns of urban vibrancy indicated by multisource urban sensing data (points-of-interest, social media check-ins, and mobile phone records) were investigated. A comprehensive urban vibrancy metric was formed by adaptively weighting these metrics. The association between urban vibrancy and demographic, economic, and built environmental factors was revealed with global regression models and local regression models. An empirical experiment was conducted in Shenzhen. The results demonstrate that four urban vibrancy metrics are all higher in the special economic zone (SEZ) and lower in non-SEZs but with different degrees of spatial aggregation. The influences of employment and road density on all vibrancy metrics are significant and positive. However, the effects of metro stations, land use mix, building footprints, and distance to district center depend on the vibrancy indicator and location. These findings unravel the commonalities and differences in urban vibrancy metrics derived from multisource urban big data and the hidden spatial dynamics of the influences of associated factors. They further suggest that urban policies should be proposed to foster vibrancy in Shenzhen therefore benefit social wellbeing and urban development in the long term. They also provide valuable insights into the reliability of urban big data-driven urban studies.