Computational Network Science: An Algorithmic Approach (Computer Science Reviews and Trends)
Format: PDF / Kindle (mobi) / ePub
The emerging field of network science represents a new style of research that can unify such traditionally-diverse fields as sociology, economics, physics, biology, and computer science. It is a powerful tool in analyzing both natural and man-made systems, using the relationships between players within these networks and between the networks themselves to gain insight into the nature of each field. Until now, studies in network science have been focused on particular relationships that require varied and sometimes-incompatible datasets, which has kept it from being a truly universal discipline.
Computational Network Science seeks to unify the methods used to analyze these diverse fields. This book provides an introduction to the field of Network Science and provides the groundwork for a computational, algorithm-based approach to network and system analysis in a new and important way. This new approach would remove the need for tedious human-based analysis of different datasets and help researchers spend more time on the qualitative aspects of network science research.
- Demystifies media hype regarding Network Science and serves as a fast-paced introduction to state-of-the-art concepts and systems related to network science
- Comprehensive coverage of Network Science algorithms, methodologies, and common problems
- Includes references to formative and updated developments in the field
- Coverage spans mathematical sociology, economics, political science, and biological networks
studies on social networks focus on static representation of nodes (i.e., humans) and edges (i.e., relationships) resulting in sociograms. By static, we mean all the relationships are assumed to occur at the same time (see Figure 5.1a). Often networks change radically as shown in progression from Figure 5.1a to c over a 36-year time span. In a weather forecasting system, weather patterns consisting of time series that are recently collected data are fed into a computer model to predict the
times. In reality, friendships and meetings are not constant. Friends come and go, and you meet different friends at different times at varying frequencies. Temporal Network Dynamics 37 Fig. 5.1. Three snapshots of a simple family network over a 36-year time span (pictures are images of a prototypical family). (a) A sociogram with a father, a mother, and a child; (b) the same sociogram after 30 years, at largest size, with the father, the mother, and the five children; (c) the same
state. To state things more formally, states of nodes can be captured by a vector xt at time t. Each node i’s probability that it is in state 1 (as opposed to being in state 0) is denoted by xt (i ) , and x0 (i ) is the state of node i at t = 0. Equation 7.1 is the update function for changes in node i’s states. M is the stochastic transition matrix, where Mij = 1/ di and di is the degree of node i. Dynamics of the voter model is summarized in Equation 7.2. When t → ∞, the state ceases to change
Externalities exist in limited settings. Consider the two-player game shown in Figure 9.2, where there are four strategies of adopting and 78 Computational Network Science: An Algorithmic Approach Fig. 9.2. An economic decision game payoff bimatrix for adopting a new technology game. avoiding changes. If a > d and b > g, each player earns a higher payoff if the player adopts; therefore, network externalities exist. The game has two Nash equilibria of (adopt, adopt) and (avoid, avoid). If
subjectively entered. This drawback inhibits its wider applicability and proliferation of his theory. 11.1 CONCLUSION NOs are emerging groups that use the Internet including social media to facilitate collaboration. This has provided serendipitous and emergent interaction among individuals that exceeds traditional organizational framework. This is a nascent area that requires attention. We envision this technology to transform information society into cohesive wholes. REFERENCES