Analyzing social network discussions of infrastructure projects: towards a bottom-up decision making environment

Community debates on infrastructure discussion networks (IDN) through social media outlets (such as Facebook and Twitter) are increasingly used as means to communicate features of infrastructure projects with communities, educate them regarding the project in different phases, and (ideally) collect their feedback. Participation of the public through online social media (and particularly through micro-blogging), referred to as 'Micro-participation', is studied by researchers. However, the lack of tools and processes to analyze seemingly chaotic public inputs of micro-participation is wasting the opportunities and becoming frustrating to both communities and official decision makers. Most of studies to this time have focused on how to build an online relationship with users to engage them in a dialogue, and more or less have ignored the social networks formed behind such dialogues. Analysis of the IDN should be the intersection of semantics and social analyses; while the former determines 'what' has been uttered, the latter reveals 'who' the utterer is. Off-the-shelf tools and commercial software systems cannot be the best answer in this regard, unless being validated and customized through adequate empirical analysis and investigation in nature of existing IDNs. There is a need for a more comprehensive analysis, especially to study the accuracy and relevance of available measures with respect to the specific nature of IDNs. This requires revisiting tools traditionally used in the domain of construction, and also benchmarking tools from other domains, and comparing their performance in a variety of IDN instances. Beyond a communication channel, IDNs contain 'context-sensitive' knowledge embedded in the apparently chaotic community debates. Similar to any other domain, the potential for applying business intelligence derived from big-data requires processing data, extracting knowledge, and providing decision makers with decision support tools. In fact, there is a potential to revise the whole theory of project management/stakeholder management form a top-down and technically-controlled process, where officials and professionals make the decisions and then communicate them with the public in a "design, decide and defend" mode/mentality; into a bottom-up and socially-savvy process, where the decisions evolve based on discussions between community and professionals. This research aims to be a starting step towards such a bottom-up system. This study offers a combination of mathematical methods and information retrieval techniques for detecting, labeling, and profiling important followers of infrastructure projects in social media, along with their communities in the mishmash of IDNs. It has also developed a framework to make a connection between the users/communities, and their ideas in the process of micro-participation. This framework supports processing big-data into the knowledge which can be efficiently used in descriptive stakeholder analysis. General or specialized online user interfaces have marked a new chapter in knowledge generation and sharing for quite a few industries such as supply chain, IT, computer programming, etc. This research was expanded on a futuristic view on the infrastructure project management in the same direction. The idea of a combined network of official and non-official decision makers, promoted by this research, requires more experiments over time to settle and stabilize. Nevertheless, methods presented in this research are applicable to any online (or offline) platform in which connectivity of people discussing construction of infrastructure is recorded, and ideas they discuss are archived. What is presented in this thesis can be used to create guidelines to decide who, and at which level, must be involved in the engagement and negotiation processes; whose concerns and the interests might have impacts on the final decisions made during the life cycle of an infrastructure project. Furthermore, it provides a tool to monitor the online social media and detect feedback/reactions with res ect to decisions made for an infrastructure project. The method presented here has the eminence of being self-organizing and emerging out of interactions of the actors from 'within' the system.