Trust Issues on Crowd-Sourcing Methods for Urban Environmental Monitoring

Introduction

Pioneering studies have been undertaken that have proven the potential worth of so-called “crowd-sourced” mobile phone data (Paulos, 2009; Bessis, 2010; Bessis et al., 2010; Sotiriadis et al., 2010; Asimakopoulou & Bessis, 2011). Several pilot studies have shown that mobile phones and mobile sensors can be used by ordinary “citizens” to gather environmental data in urban settings. Mobile sensing is still an emergent area where the costs and benefits are still rather hard to determine because it is still so early in the research cycle. Air quality in a WHO (World Health Authority) sponsored study (WHO, 2006) in Accra (Ghana), has been gathered by Taxi drivers and various pilot systems have been and are being developed now to capture urban temperature and noise levels (Maisonneuve et al., 2009). The aim is to create real-time and dynamic environmental “maps” which in turn are used to produce environmental reports, thus informing urban planning strategy and policy decision making as well as serving to monitor individual health levels and welfare (Bessis, 2010). Paulos (2009) has coined the term “citizen science” for solutions that seek to leverage collective citizen based data collection. However, participatory data collection activities of this kind and their subsequent aggregation and analysis by decision makers pose potentially serious concerns as regards security and privacy. For example, users equipped with GPS (Global Positioning Software) enabled mobile devices necessarily reveal location and contextual information as they upload data to a central data repository. Similarly, the mobile devices themselves may be exposed to risks associated with their enablement as crowd sensing devices. One concern is how best to ensure privacy whilst uploading useful information. Users may wish only to share certain directly observed data-sets to ensure their anonymity. Downloading an “app” may expose the device (hence user) to “malware” or other security infection risks. Modification of mobile devices via the installation of sensors risks the installment of so-called “Trojans”.

In a wider context, there are clearly relevant issues of “social capital” (Putnam, 2000; Hardin, 2002) thus, high-level organisational trust and reputation issues also need to be addressed if “citizen science” is to be adopted successfully to aid decision makers. Citizens need to have trust in a “civil society” and trust planners need to ensure that the data gathered by citizens is used to for the benefit of citizens and not to used merely to defend vested interests. Similarly, decision makers (such as those within e-government contexts) will need to rigorously enforce citizen data protection rights and ensure that individual rights and data privacy are maintained and not compromised by such systems. E-governance issues are raised and as with Cloud platforms organizations need to carry out effective risk assessments well before systems are enabled. That is, suitable safeguards need to be in place so as to achieve a fair balance as between the rights of citizens and the responsibilities of decision makers operating within “civil society” organisational contexts, such as e-government.

This exploratory paper seeks to offer a critique of relevant trust and security issues associated with this particular form of collective computational intelligence (Bessis, 2010). In so doing we seek to present an analysis at the social, pragmatic, semantic, syntactic and empiric levels of abstraction, whilst also presenting a balanced risks vs. rewards analysis of crowd-sourcing within a given scenario, which is typical of urban environmental monitoring and decision making contexts. We offer some potential solutions by which this exciting technology including mobile mediated data collection can best be exploited (but not misused) by decision makers operating at a local level.