The rapidly growing complex network science has presented novel approaches to complex systems modeling that were not fully foreseen even in a decade ago. It addresses the self-organization of complex network structure and its implications for system behavior, which holds significant cross-disciplinary relevance to many fields of natural and social sciences, particularly in today's highly networked social/political/economical circumstances.
Interestingly, complex network science has so far addressed either "dynamics on networks" (state transition on a network with a fixed topology) or "dynamics of networks" (topological transformation of a network with no dynamic state changes) almost separately. In many real-world complex biological and social networks, however, these two dynamics interact with each other and coevolve over the same time scales. Modeling and predicting state-topology coevolution is now recognized as one of the most significant challenges in complex network science.
In this project, the researchers will establish a generalized modeling framework that can effectively describe state-topology coevolution of complex adaptive networks and develop computational methods for automatic discovery of dynamical rules that best capture both state transition and topological transformation in the empirical data. To achieve this goal, graph rewriting systems will be used as a means of unified representation of state transition and topological transformation. Network evolution will be formulated in two parts, extraction and replacement of subnetworks. For each part, algorithms for automatic rule discovery will be explored and developed. Their effectiveness will be evaluated through application to real-world network data. This project will produce a novel theoretical framework and a computational toolkit that will transform the ways of studying the dynamics on and of complex networks.
The outcomes of this project will be disseminated via various channels and integrated in multiple educational programs at Binghamton University and other institutions. The developed algorithms and software tools will be made freely available to researchers and other professionals for their own use. The developed framework will also serve as a generalized conceptual/mathematical "language" for modeling, analyzing and discussing the dynamics of various complex systems, which will galvanize interdisciplinary discussion and collaboration across many different areas of applications. Two graduate research assistants will be supported. Members of underrepresented groups (women, minority individuals) will be particularly encouraged in the recruitment process.
The rapidly growing complex network science has presented novel approaches to complex systems modeling that were not fully foreseen even in a few decades ago. It addresses the self-organization of complex network structure and its implications for system behavior, which holds significant cross-disciplinary relevance to many fields of natural and social sciences, particularly in today's highly networked social/political/economical circumstances. Interestingly, complex network science has traditionally addressed either "dynamics on networks" (state transition on a network with a fixed topology) or "dynamics of networks" (topological transformation of a network with no dynamic state changes) almost separately. In many real-world complex biological and social networks, however, these two dynamics interact with each other and often coevolve over the same time scales. Modeling and predicting state-topology coevolution is now recognized as one of the most significant challenges in complex network science. The goals of this NSF-funded project were to establish a generalized modeling framework that could effectively describe state-topology coevolution of complex adaptive networks and to develop computational methods for automatic discovery of dynamical rules that best capture both state transition and topological transformation in empirical data. To achieve these goals, we used graph rewriting systems as a means of unified representation of state transition and topological transformation. We formulated network evolution in two parts, extraction and replacement of subnetworks. For each part, we explored and developed algorithms for automatic rule discovery. Their effectiveness was evaluated through application to simulated and real-world network data. We also developed methodologies of collecting real-world temporal network data from online resources, and expanded the domains of applications of the adaptive network framework to the modeling of social and organizational network evolution, collective behavior of heterogeneous swarms, joint technological systems, and social/cultural diffusion. This project has produced a novel theoretical framework and a computational toolkit that are expected to become the basis of transformational ways of studying the coevolution of dynamics on and of complex networks in the coming years. The developed framework has served as a generalized conceptual/mathematical "language" for modeling, analyzing and discussing the dynamics of various complex systems, which has galvanized interdisciplinary discussion and collaboration across many different areas of applications. Its outcomes have been disseminated via various channels, including 5 peer-reviewed journal articles (1 more currently under review and 3 more in preparation), 12 published conference papers/abstracts (including two that won Best Student Paper and Best Presentation awards), 30 presentations, 1 Master’s thesis, and 1 unpublished mathematical proof. Several symposia were also organized on adaptive networks. The research outcomes have been integrated into multiple educational programs at Binghamton University and other institutions. The developed algorithms and software tools are made freely available to researchers and other professionals for their own use. Finally, this project helped create the NetSci High educational outreach program that connects high school students and teachers to regional research labs, offering them an opportunity to work on a year-long research project on network science. This program has grown dramatically over the course of this project, to a highly recognized outreach program at international levels. One of the published journal articles was co-authored by high school students and teachers who participated in this program, which attracted a lot of international media attention.
