Exploratory Analysis in Dynamic Social Networks

Social interactions within networks comprise an increasing event nowadays. Different aspects of societies and competitive markets, such as Social Medias through the Internet and Telecommunications environments hold a high correlation with the social network analysis methodology. By understanding these social structures and its interactions might be possible to realize how individuals and consumers relate each other and hence predict further social structures in the future. However, most of the current social network analysis projects are in relation to static structures, not considering how the social network evolves over the time. The dynamic approach can points out new perspectives in terms of social network analysis, including prediction and simulations scenarios. In order to perceive the social network relations over the time is crucial to collect the distinct snapshots of the social structure, understanding not just how the social members relate each other but in addition to that how this relationships evolves over the time. Measures in relation to social network describe nodes and links by static metrics, depicting its strength, its overall distances to the other related nodes, and its amounts of connections, among others. This dynamic approach makes possible to create a historical data, quite usual for predictive modeling. As such, new social network measures and algorithms should be created in order to describe dynamic features assigned to social structures over the time. Chapter 1 gives a brief overview of social network, such as its characteristics and how to visualize it. Chapter 2 discusses the general practice of building pervasive online social networks using real-world services and presents various projects that focus on the online social networking of the physical world. Chapter 3 presents a semantic model, non probabilistic and predictive, for the decisional analysis of professional and institutional social networks. Chapter 4 describes a novel approach towards the visualization and analysis of network dynamics. The goal is to handle network data and visualization in ways that explicitly deal with its time-based nature while simultaneously assessing nodal- and macro- network dynamics. Chapter 5 shows how the network topology affects epidemic diffusion and cascade dynamics. It also clarifies the largest maximum eigenvalue of the adjacency matrix of the network is the key index to estimate the properties of networks. Chapter 6, discusses the behaviors of malware propagation in online social networks using analytical models and conducts some simulations. Chapter 7 analyzes the Google Analytics data from multiple web companies. These web services all positioned themselves as social network sites within a relatively short time from their establishment. Several important measures available on Google Analytics were discussed with experienced practitioners to be identified for data collection. Chapter 8 explores the effects of spatiality and the direction of information transmission over cooperation dynamics. The model is based on the well-known "selfish herd" concept, and assumes that cultural and biological dynamics are driven by natural selection of the phenotypes. This model allows us to study the differences between the dynamics of cooperative, group-forming individuals subject to a selective pressure (predation). Chapter 9 does the survey about the types of private information and privacy threats on social network graphs, and then introduce the methods to protect the private information. Chapter 10 reviews the e-Government development processes in Canada, Malaysia, Australia, Singapore and Korea in order to identify the factors that contribute to the successful development of e-Government therein.