Citizen centric stakeholder theory: sentiment and behavior analyses in social media

In a relatively short period of time, social media has gained significant importance as a mass communication and public engagement tool. Rapid dissemination of information through social media platforms such as Facebook and Twitter, provides individuals and organizations with the ability to broadcast their message to a wide audience instantly and directly. Hence, we now witness almost all public and private organizations with a presence on popular social networks. However, unlike conventional media, social media allows users to wield tremendous influence over popular discussion topics and events. While in traditional media, the narrative is built by a group of established influencers, discussions remain fluid and decentralized on social networks. Although some users do wield more sway than others, all citizens utilizing the platform are stakeholders who have to be recognized and managed if any entity wishes to use these platforms to carry out their mission. The primary goal of this research is to perform analysis of the social media data to gain insights and to better understand the nature of the discourse taking place on popular social media platforms and how sentiment, keywords and users impact these discussions. Such understanding is important to enable prime movers of social media to accurately identify and engage their stakeholders, enabling them to foster more meaningful communication with online social network users. Towards this end, this dissertation has made the following contributions: First, it has developed a framework to analyze social media usage by adopting the time tested stakeholder theory for social media. It has proposed a Citizen Centric Stakeholder Theory for Social Media, where all users of social media platforms are treated as stakeholders. Second, it has utilized the above framework to investigate the nature and characteristics of social media usage by citizens from around the world. It has evaluated this participation through sentiment and user behavior analysis. As a case study, this dissertation has applied this citizen centric stakeholder theory to analyze Twitter data gathered during US Presidential Elections of November 2016 and UK General Elections of June 2017. Third, it has evaluated the social media communication of public sector organizations operating in the Northeast US to understand their communication strategy in terms of sentiment and message content. Fourth, it has studied the subjectivity and polarity of Twitter data with respect to the location of Tweets. Finally, a web-based system for geospatial visualization of current sentiments associated with a hashtag or word in Twitter messages has been developed.Ph.D.Includes bibliographical referencesby Ussama Yaqu