Network theory

About: Network theory is a research topic. Over the lifetime, 2257 publications have been published within this topic receiving 109864 citations.

Parameterized Centrality for Network Analysis

Abstract: Bonacich centrality measures the number of attenuated paths between nodes in a network We use this metric to study network structure, specifically, to rank nodes and find community structure of the network To this end we extend the modularity-maximization method for community detection to use this centrality metric as a measure of node connectivity Bonacich centrality contains a tunable parameter that sets the length scale of interactions By studying how rankings and discovered communities change when this parameter is varied allows us to identify globally important nodes and structures We apply the proposed method to several benchmark networks and show that it leads to better insight into network structure than earlier methods

Vehicle-Level Electromagnetic Compatibility Prediction Based on Multi-Port Network Theory

Abstract: This paper proposes a new methodology based on the multi-port network theory to predict the vehicle-level electromagnetic compatibility performance. The original EMC problem is firstly converted to a network by separating the electrical large structures and electrical small components. The impedance is proposed to describe the coupling process of network to eliminate the influence of port impedance on network. Based on this network model, the relationship between the exciting sources and the sensitive components is set up using the multi-port network theory. Furthermore, some application problems, such as measurement of parameters, are also discussed. After validated by a bench test, this methodology for vehicle level electromagnetic compatibility was further applied to predict and improve the low frequency radiated emission of an electric vehicle. The application results show that it can be used to predict electromagnetic interference and analyze the main exciting source satisfactorily.

Extensions of maximum concurrent flow to identify hierarchical community structure and hubs in networks

Abstract: This research has lead to the development of three new tools for social network analysis that are extensions of the maximum concurrent flow problem. One is the Maximum Concurrent Flow (MCF) cut algorithm that identifies the hierarchical community structure of a network by iteratively solving the maximum concurrent flow problem for the least-dense component. A second is a new measure of accuracy for comparing two partitionings of a network. It is based upon the inter-community and intra-community classification of edges rather than the problematic vertex counting that has been universally employed for this purpose. The third tool is a new, canonical centrality measure based upon the hierarchical maximum concurrent flow problem that is more stable than the currently-used betweenness centrality measure and is useful in identifying hub-type vertices within networks that conduct higher-levels of flow between other vertices. The new centrality measure can be a tool to improve research that involves social network analysis and other fields involved with the analysis of network flows. The proposed “flowthrough” centrality measure overcomes the problem of overstating or understating the roles that significant actors play in social networks. Rigorous experiments have found this measure to be more stable than the traditional betweenness centrality measurement used in social network analysis when knowledge of the network topology is incomplete or in transition. Perturbations do not alter the flowthrough centrality values of vertices based upon all realized flow as much as they do betweenness centrality values based upon geodesics only. It is canonical in that it is determined from a natural and standardized flow applicable to all networks. The major benefits of this study are the development of more reliable tools for the identification of the hierarchical community structure of networks, a new measure of accuracy for comparing partitionings of networks, a new and more stable flowthrough centrality measurement that can identify hub vertices, and the evidence for that reliability.

Network Detection Theory

Abstract: Despite the breadth of modern network theory, it can be difficult to apply its results to the task of uncovering terrorist networks: the most useful network analyses are often low-tech, link-following approaches. In the traditional military domain, detection theory has a long history of finding stealthy targets such as submarines. We demonstrate how the detection theory framework leads to a variety of network analysis questions. Some solutions to these leverage existing theory; others require novel techniques – but in each case the solutions contribute to a principled methodology for solving network detection problems. This endeavor is difficult, and the work here represents only a beginning. However, the required mathematics is interesting, being the synthesis of two fields with little common history.