Fabrikant Salvini

Fakultäten » Mathematisch-naturwissenschaftliche Fakultät » Geographisches Institut » Geographic Information Science » Prof. Dr. Sara Irina Fabrikant » Fabrikant Salvini

Completed research project

Title / Titel Relational urban studies – A triangulation of computational, semantic, and social network analyses, coupled with spatialization methods
PDF Abstract (PDF, 14 KB)
Summary / Zusammenfassung INTRODUCTION
Castells (1996) identifies the “space of flows” as a new spatial form with an increasingly dominant role in today’s network society. At the urban level, the space of flows is constituted by the exchanges of people, goods, and information between cities. A flow network can be seen as an expression of the integration of cities in an urban system. This functional network shapes and transforms the linked cities. However, the identification and modeling of a functional network and the measurement of flows within it is one of the key challenges in this research domain within economic geography.
One pressing research issue is the identification of appropriate flow data to model functional networks. For example, previous research has typically employed the interlocking of multinational enterprises, or of advanced producer services, or even flight passenger volumes between world cities. While interlocking network of enterprises might serve as a theoretical basis for modeling the functional network of economical relations of world cities, the functional meaning of flight passenger volumes between cities is less obvious.
A major research challenge for world city research is to theoretically and methodologically ground the employed relational datasets for establishing a functional network. Short et al. (1996) speak in this context of the dirty little secret of world cites research.

This PhD project aims at tackling this particular problem by answering the following research questions:
- How appropriate are openly available massively “crowd-sourced semantic databases” (i.e., Wikipedia) for the exploration and analysis of urban systems with a relational perspective?
- How can a triangulation of established methods from computational linguistics, GIScience and computer science transfer to and enrich the sets of methods commonly used in the area of world and global city research within economic geography?

We aim at answering these generic methodological research questions by applying our new interdisciplinary approach to a specific case study on the Swiss urban system with the following content driven research questions:
Can we gain new insights about the Swiss urban system with the proposed approach? More specifically, how embedded are Swiss cities in an international urban system, and where are these Swiss cities located relationally in a respective functional international city network? Is there a concentration of the emergent activities with complex internal relations in one or more mega-city regions on a national level? Who are the actors in the Swiss city system?

This research project at the intersection of economical geography and geographic information visualization and analysis within GIScience investigates the potential of massively „crowd-sourced” semi-structured text databases“ (e.g., Wikipedia) for the systematic analysis of urban systems, and in particular the space of flows of within these systems. Our proposed methodological framework is put to a rigorous test in a case study to uncover the latent structure of the Swiss urban system and its embedment within an international urban system. Specifically, we aim at uncovering the relational structure of the Swiss cities by analyzing Wikipedia articles that mention these cities. This analysis follows a two-pronged approach. Firstly, we employ graph-theoretic methods to reveal the web link structure of the articles of interest, and secondly we identify relationships between cities through a computational semantic analysis of the content in these respective articles.
The semantic content is explored and analyzed with the integration of theoretically sound quantitative methods from social science (i.e., social network analysis) together with computational approaches from GIScience and computer science (i.e., spatial analysis, geovisualization, and graph drawing). The results are systematical validated.

Key outputs are expected in three areas. First, a generic methodological framework which is potentially applicable to any urban system on any geographic scale is developed and validated. Second, we test the feasibility for employing massively growing and openly available crowd-sourced databases that might be one solution to solve the “dirty little secret” problem (i.e., the lack of available functional differentiated relational data) for world cities research. In doing so, we hope to uncover social, economical, political and cultural relations between cities. Third, by integrating robust and complementary methods from separate research fields we hope to contribute to the extension of sound quantitative methods employed in economic geography that will hopefully bring new and more robust results to uncover the urban structure in general, and of the Swiss urban system in particular.
Publications / Publikationen Gnos, A.U., Salvini, M.M., Fabrikant, S.I. (2011). Cognitively Plausible Visualization of Network Data. Proceedings, 25th International Cartographic Conference, International Cartographic Association, Jul. 3-8, 2011, Paris, France (refereed on extended abstract).

Fabrikant, S.I., Salvini, M.M. (2011). Charting the ICA World of Cartography 1999–2009. Proceedings, 25th International Cartographic Conference, International Cartographic Association, Jul. 3-8, 2011, Paris, France (refereed on extended abstract).

Gnos, A. (2010). Visualisierung von Stadtnetzwerken. Unpublished MSc thesis, Department of Geography, University of Zurich. (Advisors: M. Salvini, S.I. Fabrikant).

Keywords / Suchbegriffe world cities theory, social network analysis, spatialization
Project leadership and contacts /
Projektleitung und Kontakte
Prof. Sara Irina Fabrikant, PhD (Project Leader)
Marco Salvini, MSc (Project Leader)
Funding source(s) /
Unterstützt durch
Universität Zürich (position pursuing an academic career)
In collaboration with /
In Zusammenarbeit mit
Prof. Dr. Céline Rozenblat, Université de Lausanne, Lausanne Switzerland
Duration of Project / Projektdauer Jan 2010 to Jan 2013