Cityware: Urban Computing to Bridge Online and Real-World Social Networks

Cityware: Urban Computing to Bridge Online and Real-World Social Networks

Vassilis Kostakos (University of Bath, UK) and Eamonn O’Neill (University of Bath, UK)
DOI: 10.4018/978-1-60566-152-0.ch013
OnDemand PDF Download:
No Current Special Offers


In this paper, we describe a platform that enables us to systematically study online social networks alongside their real-world counterparts. Our system, entitled Cityware, merges users’ online social data, made available through Facebook, with mobility traces captured via Bluetooth scanning. Furthermore, our system enables users to contribute their own mobility traces, thus allowing users to form and participate in a community. In addition to describing Cityware’s architecture, we discuss the type of data we are collecting, and the analyses our platform enables, as well as users’ reactions and thoughts.
Chapter Preview


The formalised study of network graphs is considered to have begun by Euler’s famous solution to the Seven Bridges of Königsberg problem in 1736 (Biggs et al., 1986). In his solution, Euler represented the four landmasses and seven bridges of Königsberg, now Kaliningrad, as four nodes and seven links respectively. Thus, he was able to prove that no route crosses each bridge only once. Graph theory has greatly advanced every since, mostly focusing on mathematical proofs and theorems on graph topology, trees and cycles.

While graphs have been used to explore relationships between social entities for over a century, it was not until the 1950’s that this became a systematic, and ultimately scientific process. Some of the first studies to engage in social network analysis are the kinship studies of Elizabeth Bott (Bott, 1957) and the urbanisation studies pioneered by Max Gluckman in Zambia (Gluckman & Aronoff, 1976). Similarly, Granovetter’s work (1973) lay the foundations for the small world hypothesis, suggesting that everyone is within six degrees of separation, while Wellman’s work gave some evidence of how large-scale social changes have affected the nature of personal communities and the support they provide (1979). Since then, social network analysis has moved from being a suggestive metaphor to becoming an analytic approach, with its own theories and research methods. In the 1970’s, Freeman developed a multitude of metrics for analysing social and communication networks (e.g. 2004), thus boosting commercial interest in the area due to companies aiming to optimise their procedures and operations. In the last decade, the identification of mathematical principles such as the small-world and scaling phenomena (Barabasi & Albert, 1999; Watts & Strogatz, 1998), underpinning many natural and man-made systems, have sparked further interest in the study of networks.

The systems design community has also been interested in the study of social networks as well as online social networks. Typical research topics in the area include the effect of social engagement on behaviour (e.g. Millen & Patterson, 2002), the issue of identity and projected identity (Lee & Nass, 2003), as well as the design of socio-technical systems (Herrmann et al., 2004). The recent proliferation of online social networking system such as Facebook, Dodgeball and MySpace, has provided researchers with platforms for carrying out research into online social behaviour, and a journal devoted to this topic ( In the Urban Computing domain, such studies have looked at the effect of social incentives and contextual information on the use of public transportation (Booher et al., 2007), the relationship between users’ online profiles and their online behaviour (Lampe et al., 2007), the various trust issues that emerge from using such systems (Riegelsberber & Vasalou, 2007), how such systems can help strengthen neighbourhoods (Foth, 2006), and the development of systematic grounds to base our designs (Kostakos et al., 2006).

To make inferences from online behaviour datasets, researchers still have to collect data from the real world and relate it to the online data. Thus, while social networking websites make it easy to capture large amounts of data, researchers still need to employ interviews, focus groups, questionnaires, or any other method that enables them to relate online with real world data.

Key Terms in this Chapter

Bluetooth Identifier: A unique 12-digit hexadecimal number used by Bluetooth components for identification.

Social Network: A structure that represents social relationships. The strutter typically consists of nodes and links between the nodes, and the nodes represent people while the links represent a specific type of relationship such as friendship, marriage, or financial relationship.

Aggregate Patterns of [Behaviour/Encounter/Diffusion]: On an individual level each person behaves in distinct and unique ways, having specific objectives in mind. Yet, when analysed at an aggregate level, communities and cities exhibit non-random patterns that emerge from the combination of each distinct person’s activities. Such patterns are known as aggregate patterns, and can describe how people encounter each other, or how information is diffused and spread through the community.

Urban Computing: A research field focused on the development of computer systems that are to be used in urban space. Typically, such systems entail fixed, mobile and embedded components.

Massively Distributed System: A real-time computer system with large numbers of physical and logical components spanning great geographic distances.

Complete Chapter List

Search this Book: