Computational Approaches for Urban Environments

1 downloads 0 Views 138KB Size Report
Published by Taylor & Francis, LLC. Computational Approaches for. Urban Environments. Marco Helbich, Jamal Jokar. Arsanjani, and Michael Leitner, eds.
The AAG Review OF BOOKS

Computational Approaches for Urban Environments Marco Helbich, Jamal Jokar Arsanjani, and Michael Leitner, eds. Basel, Switzerland: Springer International, 2015. 395 pp., maps, figures, tables, index, bibliography. $179.00 cloth (ISBN 978-3-319-11468-2). Reviewed by Xiaohui Liu, Department of Geography & Geology, University of Southern Mississippi, Hattiesburg, MS. Being dynamic, complex, and vibrant in nature, cities never cease to urbanize, which continuously changes their appearance and subcomponents. The desire to understand the spatial and spatiotemporal patterns and processes of each component of urban environment, which consists of urban planning, housing and real estate, urban transportation, retail industry, and so on, drives applications using cutting-edge computational approaches to examine urban environments. Seen from this angle, the integration of advanced computational methodologies and urban science is an inevitable outcome of urban development; hence, this book came into existence. Due to inherent distinctions among subfields of urban research, each chapter mainly focuses on one particular theme. Beginning with spatial planning and decision making, which is a traditionally predominant research focus, the book touches on popular urban theories, such as fractal analysis and spatial pattern discovery. Unlike conventional urban analysis, Martin Behnisch and Alfred Ultsch adopt machine learning and data mining techniques for spatial pattern discovery, which demonstrates well how data mining and knowledge discovery can be applied to a spatial data set. It is also worth mentioning that the approach does not end with a data

cluster; rather, a cluster could generate new knowledge and be of practical use. Julian Hagenauer introduces a novel approach that combines contextual neural gas, topology learning, and graph clustering to cluster spatial data, which is proven to be effective based on a synthetic and a real-world data set, respectively. Housing and real estate have always been hot topics worldwide; thus, three chapters are dedicated to these topics. Each chapter concentrates on a computational model or theory in the context of its study. As a result of investigating homes in urban areas, Alexander Razen et al. propose a Bayesian model in addition to a multilevel structured additive regression model. Shipeng Sun and Steven Manson formulate an agent-based model to explain how common housing search behavior leads to intraurban migration. Timothy Rosner and Kevin Curtin develop a livability capability index to measure social and economic characteristics of the built environment. In line with the theme of the book, these novel computation-intensive approaches enumerate new application possibilities by renovating existing models or approaches on current social or economic issues, which will doubtlessly inspire future researchers and bring urban environment research to the next level. The next few chapters focus on urban research, transportation, and mobility. Not surprisingly, Global Positioning System (GPS) data serves as the main data set for location tracking and traffic simulation to guarantee the effectiveness of space–time models within diverse contexts. Godwin Yeboah et al. propose a space analytical approach to compare route choice preferences of participants by comparing collected GPS data to official cycling network data of British cities. This study facilitates the

The AAG Review of Books 4(3) 2016, pp. 148–149. doi: 10.1080/2325548X.2016.1187499. ©2016 by American Association of Geographers. Published by Taylor & Francis, LLC.

collection and analysis of detailed bicyclists’ route choices and provides empirical evidence for understanding utility cycling behaviors. Furthermore, Rashid Waraich et al. model individual traffic, which leads to increased computational burden, and thus, have to use parallel computing techniques to improve computational performance. Both the chapters involve transportation and mobility analysis at an individual level, which is bound to generate extensive data that call for efficient approaches to solve computing-intensive problems. Therefore, this research well represents the trend of computational approach for urban study, and definitely sheds light on related topics in this field. Remote sensing and GIScience have been inseparable in various subfields; urban environment research, as part of GIScience, is no exception. The utilization of remote sensing techniques, including traditional change detection, urbanization simulation, and airborne hyperspectral and light detection and ranging (LiDAR), adds significantly to the current research agenda on cities, and produces a situation where urban maps have even finer resolution, urban changes obtain higher accuracy, and urban models are more sophisticated. Integrating remote sensing with socioeconomic data in Junmei Tang’s implementation of urban land use change modeling is proven to have improved the accuracy, which indicates that socioeconomic data supplements remote sensing data in urban studies given the fact that the urban environment has humanintervened landscapes. With a focus on urban sensing, social network, and social media, the last part of the volume keeps abreast of current research highlights, and presents studies on social networks to facilitate understanding of cultural, technological, and economic factors shaping human dynamics. Yaoli Wang et al. leverage social data from call data records

to explore established contacts, and thus facilitate our understanding of social relationships embedded in urban physical space. This points out the necessity to deepen research on the interaction of social network analysis and urban planning. Emily Schnebele et al. propose a novel approach to use unconventional, nonauthoritative data in transportation infrastructure assessment and emergency evacuations. In the two scenarios, nonauthoritative data, as a complement to official data, provide valuable, near real-time, and on-the-ground information during disasters. This research exhibits well the interdisciplinary research nature that combines emergency management and disaster mitigation, remote sensing, geospatial analysis, big data techniques, and social media applications. Urban sensing, volunteered geographic information, and social media together provide novel perspectives and enrich the current research agenda on cities. This rare and highquality collection of articles represents the combination of the previously mentioned research topics. If there is anything to critique in this volume, it is the limited focus of articles given the great variety of studies that might investigate aspects of cities using innovative computational approaches. Should more related articles be collected, a follow-up volume would be warmly received. This volume will be of interest to broad groups of readers. From an interdisciplinary perspective, it perfectly blends urban research with GIScience and computational science, and it brings out methodological advances as for how to use novel data set and employ geospatial technologies in urban-related studies. Rapid discipline construction and deeper disciplines’ integration provide urban research with valuable and promising tools and wide application prospects. In this sense, this collection of chapters should be a mandatory reading for those working on urban studies who would like to explore an increasing urbanization through fast updating technologies.

SUMMER 2016 149