Showing papers in "Physical Review E in 2004"
TL;DR: It is demonstrated that the algorithms proposed are highly effective at discovering community structure in both computer-generated and real-world network data, and can be used to shed light on the sometimes dauntingly complex structure of networked systems.
Abstract: We propose and study a set of algorithms for discovering community structure in networks-natural divisions of network nodes into densely connected subgroups. Our algorithms all share two definitive features: first, they involve iterative removal of edges from the network to split it into communities, the edges removed being identified using any one of a number of possible "betweenness" measures, and second, these measures are, crucially, recalculated after each removal. We also propose a measure for the strength of the community structure found by our algorithms, which gives us an objective metric for choosing the number of communities into which a network should be divided. We demonstrate that our algorithms are highly effective at discovering community structure in both computer-generated and real-world network data, and show how they can be used to shed light on the sometimes dauntingly complex structure of networked systems.
12,882 citations
TL;DR: A hierarchical agglomeration algorithm for detecting community structure which is faster than many competing algorithms: its running time on a network with n vertices and m edges is O (md log n) where d is the depth of the dendrogram describing the community structure.
Abstract: The discovery and analysis of community structure in networks is a topic of considerable recent interest within the physics community, but most methods proposed so far are unsuitable for very large networks because of their computational cost. Here we present a hierarchical agglomeration algorithm for detecting community structure which is faster than many competing algorithms: its running time on a network with n vertices and m edges is O (md log n) where d is the depth of the dendrogram describing the community structure. Many real-world networks are sparse and hierarchical, with m approximately n and d approximately log n, in which case our algorithm runs in essentially linear time, O (n log(2) n). As an example of the application of this algorithm we use it to analyze a network of items for sale on the web site of a large on-line retailer, items in the network being linked if they are frequently purchased by the same buyer. The network has more than 400 000 vertices and 2 x 10(6) edges. We show that our algorithm can extract meaningful communities from this network, revealing large-scale patterns present in the purchasing habits of customers.
6,599 citations
TL;DR: An algorithm is described which gives excellent results when tested on both computer-generated and real-world networks and is much faster, typically thousands of times faster, than previous algorithms.
Abstract: Many networks display community structure--groups of vertices within which connections are dense but between which they are sparser--and sensitive computer algorithms have in recent years been developed for detecting this structure. These algorithms, however, are computationally demanding, which limits their application to small networks. Here we describe an algorithm which gives excellent results when tested on both computer-generated and real-world networks and is much faster, typically thousands of times faster, than previous algorithms. We give several example applications, including one to a collaboration network of more than 50,000 physicists.
5,127 citations
TL;DR: Two classes of improved estimators for mutual information M(X,Y), from samples of random points distributed according to some joint probability density mu(x,y), based on entropy estimates from k -nearest neighbor distances are presented.
Abstract: We present two classes of improved estimators for mutual information M(X,Y), from samples of random points distributed according to some joint probability density mu(x,y). In contrast to conventional estimators based on binnings, they are based on entropy estimates from k -nearest neighbor distances. This means that they are data efficient (with k=1 we resolve structures down to the smallest possible scales), adaptive (the resolution is higher where data are more numerous), and have minimal bias. Indeed, the bias of the underlying entropy estimates is mainly due to nonuniformity of the density at the smallest resolved scale, giving typically systematic errors which scale as functions of k/N for N points. Numerically, we find that both families become exact for independent distributions, i.e. the estimator M(X,Y) vanishes (up to statistical fluctuations) if mu(x,y)=mu(x)mu(y). This holds for all tested marginal distributions and for all dimensions of x and y. In addition, we give estimators for redundancies between more than two random variables. We compare our algorithms in detail with existing algorithms. Finally, we demonstrate the usefulness of our estimators for assessing the actual independence of components obtained from independent component analysis (ICA), for improving ICA, and for estimating the reliability of blind source separation.
3,224 citations
TL;DR: It is pointed out that weighted networks can in many cases be analyzed using a simple mapping from a weighted network to an unweighted multigraph, allowing us to apply standard techniques for unweighting graphs to weighted ones as well.
Abstract: The connections in many networks are not merely binary entities, either present or not, but have associated weights that record their strengths relative to one another. Recent studies of networks have, by and large, steered clear of such weighted networks, which are often perceived as being harder to analyze than their unweighted counterparts. Here we point out that weighted networks can in many cases be analyzed using a simple mapping from a weighted network to an unweighted multigraph, allowing us to apply standard techniques for unweighted graphs to weighted ones as well. We give a number of examples of the method, including an algorithm for detecting community structure in weighted networks and a simple proof of the maximum-flow--minimum-cut theorem.
2,386 citations
TL;DR: In this article, an improved method to retrieve the effective constitutive parameters (permittivity and permeability) of a slab of metamaterial from the measurement of S parameters is proposed.
Abstract: We propose an improved method to retrieve the effective constitutive parameters (permittivity and permeability) of a slab of metamaterial from the measurement of S parameters. Improvements over existing methods include the determination of the first boundary and the thickness of the effective slab, the selection of the correct sign of effective impedance, and a mathematical method to choose the correct branch of the real part of the refractive index. The sensitivity of the effective constitutive parameters to the accuracy of the S parameters is also discussed. The method has been applied to various metamaterials and the successful retrieval results prove its effectiveness and robustness.
1,941 citations
TL;DR: The power grid is robust to most perturbations, yet disturbances affecting key transmission substations greatly reduce its ability to function, and it is emphasized that the global properties of the underlying network must be understood as they greatly affect local behavior.
Abstract: The magnitude of the August 2003 blackout affecting the United States has put the challenges of energy transmission and distribution into limelight. Despite all the interest and concerted effort, the complexity and interconnectivity of the electric infrastructure precluded us for a long time from understanding why certain events happened. In this paper we study the power grid from a network perspective and determine its ability to transfer power between generators and consumers when certain nodes are disrupted. We find that the power grid is robust to most perturbations, yet disturbances affecting key transmision substations greatly reduce its ability to function. We emphasize that the global properties of the underlying network must be understood as they greatly affect local behavior.
1,364 citations
TL;DR: An alternative derivation of passive imaging of the ballistic wave that is not based on normal modes is presented, showing that the global requirement of the equipartitioning of normal modes can be relaxed to the local requirement that the scattered waves propagate on average isotropically near the receivers.
Abstract: The Green's function of waves that propagate between two receivers can be found by cross-correlating multiply scattered waves recorded at these receivers. This technique obviates the need for a source at one of these locations, and is therefore called "passive imaging." This principle has been explained by assuming that the normal modes of the system are uncorrelated and that all carry the same amount of energy (equipartitioning). Here I present an alternative derivation of passive imaging of the ballistic wave that is not based on normal modes. The derivation is valid for scalar waves in three dimensions, and for elastic surface waves. Passive imaging of the ballistic wave is based on the destructive interference of waves radiated from scatterers away from the receiver line, and the constructive interference of waves radiated from secondary sources near the receiver line. The derivation presented here shows that the global requirement of the equipartitioning of normal modes can be relaxed to the local requirement that the scattered waves propagate on average isotropically near the receivers.
1,089 citations
TL;DR: It is shown that the breakdown of a single node is sufficient to collapse the efficiency of the entire system if the node is among the ones with largest load.
Abstract: Large but rare cascades triggered by small initial shocks are present in most of the infrastructure networks. Here we present a simple model for cascading failures based on the dynamical redistribution of the flow on the network. We show that the breakdown of a single node is sufficient to collapse the efficiency of the entire system if the node is among the ones with largest load. This is particularly important for real-world networks with a highly hetereogeneous distribution of loads as the Internet and electrical power grids.
1,071 citations
TL;DR: The response of a model microelectrochemical system to a time-dependent applied voltage is analyzed, including electrochemistry, colloidal science, and microfluidics, including surface conduction, multicomponent electrolytes, and Faradaic processes.
Abstract: The response of a model microelectrochemical system to a time-dependent applied voltage is analyzed. The article begins with a fresh historical review including electrochemistry, colloidal science, and microfluidics. The model problem consists of a symmetric binary electrolyte between parallel-plate blocking electrodes, which suddenly apply a voltage. Compact Stern layers on the electrodes are also taken into account. The Nernst-Planck-Poisson equations are first linearized and solved by Laplace transforms for small voltages, and numerical solutions are obtained for large voltages. The "weakly nonlinear" limit of thin double layers is then analyzed by matched asymptotic expansions in the small parameter epsilon= lambdaD/L, where lambdaD is the screening length and L the electrode separation. At leading order, the system initially behaves like an RC circuit with a response time of lambdaDL/D (not lambdaD2/D), where D is the ionic diffusivity, but nonlinearity violates this common picture and introduces multiple time scales. The charging process slows down, and neutral-salt adsorption by the diffuse part of the double layer couples to bulk diffusion at the time scale, L2/D. In the "strongly nonlinear" regime (controlled by a dimensionless parameter resembling the Dukhin number), this effect produces bulk concentration gradients, and, at very large voltages, transient space charge. The article concludes with an overview of more general situations involving surface conduction, multicomponent electrolytes, and Faradaic processes.
938 citations
TL;DR: The existence of acoustic metamaterial, in which both the effective density and bulk modulus are simultaneously negative, in the true and strict sense of an effective medium, is shown.
Abstract: We show here the existence of acoustic metamaterial, in which both the effective density and bulk modulus are simultaneously negative, in the true and strict sense of an effective medium. Our double-negative acoustic system is an acoustic analogue of Veselago's medium in electromagnetism, and shares many unique consequences, such as negative refractive index. The double negativity in acoustics is derived from low-frequency resonances, as in the case of electromagnetism, but the negative density and modulus are derived from a single resonance structure as distinct from electromagnetism in which the negative permeability and negative permittivity originates from different resonance mechanisms.
TL;DR: It is shown both numerically and analytically that random graphs and scale-free networks have modularity and it is argued that this fact must be taken into consideration to define statistically significant modularity in complex networks.
Abstract: The mechanisms by which modularity emerges in complex networks are not well understood but recent reports have suggested that modularity may arise from evolutionary selection. We show that finding the modularity of a network is analogous to finding the ground-state energy of a spin system. Moreover, we demonstrate that, due to fluctuations, stochastic network models give rise to modular networks. Specifically, we show both numerically and analytically that random graphs and scale-free networks have modularity. We argue that this fact must be taken into consideration to define statistically significant modularity in complex networks.
TL;DR: The mean-field equations characterizing the dynamics of a rumor process that takes place on top of complex heterogeneous networks are derived and the time profiles for several quantities are numerically computed, which allows us to distinguish among different variants of rumor spreading algorithms.
Abstract: We derive the mean-field equations characterizing the dynamics of a rumor process that takes place on top of complex heterogeneous networks. These equations are solved numerically by means of a stochastic approach. First, we present analytical and Monte Carlo calculations for homogeneous networks and compare the results with those obtained by the numerical method. Then, we study the spreading process in detail for random scale-free networks. The time profiles for several quantities are numerically computed, which allows us to distinguish among different variants of rumor spreading algorithms. Our conclusions are directed to possible applications in replicated database maintenance, peer-to-peer communication networks, and social spreading phenomena.
TL;DR: Analytical results are derived, showing that the proposed class of models of social network formation reproduces the main statistical characteristics of real social networks: large clustering coefficient, positive degree correlations, and the emergence of a hierarchy of communities.
Abstract: We propose a class of models of social network formation based on a mathematical abstraction of the concept of social distance. Social distance attachment is represented by the tendency of peers to establish acquaintances via a decreasing function of the relative distance in a representative social space. We derive analytical results (corroborated by extensive numerical simulations), showing that the model reproduces the main statistical characteristics of real social networks: large clustering coefficient, positive degree correlations, and the emergence of a hierarchy of communities. The model is confronted with the social network formed by people that shares confidential information using the Pretty Good Privacy (PGP) encryption algorithm, the so-called web of trust of PGP.
TL;DR: A continuum field theory approach is presented for modeling elastic and plastic deformation, free surfaces, and multiple crystal orientations in nonequilibrium processing phenomena.
Abstract: A continuum field theory approach is presented for modeling elastic and plastic deformation, free surfaces, and multiple crystal orientations in nonequilibrium processing phenomena. Many basic properties of the model are calculated analytically, and numerical simulations are presented for a number of important applications including, epitaxial growth, material hardness, grain growth, reconstructive phase transitions, and crack propagation.
TL;DR: This work studies the family of network models derived by requiring the expected properties of a graph ensemble to match a given set of measurements of a real-world network, while maximizing the entropy of the ensemble.
Abstract: We study the family of network models derived by requiring the expected properties of a graph ensemble to match a given set of measurements of a real-world network, while maximizing the entropy of the ensemble. Models of this type play the same role in the study of networks as is played by the Boltzmann distribution in classical statistical mechanics; they offer the best prediction of network properties subject to the constraints imposed by a given set of observations. We give exact solutions of models within this class that incorporate arbitrary degree distributions and arbitrary but independent edge probabilities. We also discuss some more complex examples with correlated edges that can be solved approximately or exactly by adapting various familiar methods, including mean-field theory, perturbation theory, and saddle-point expansions.
TL;DR: This investigation details the properties of a passive, dispersive metamaterial that is matched to free space and has an index of refraction equal to zero, and shows that in both the source and scattering configurations the electromagnetic fields in a matched zero-index medium take on a static character in space, yet remain dynamic in time.
Abstract: Planar metamaterials that exhibit a zero index of refraction have been realized experimentally by several research groups. Their existence stimulated the present investigation, which details the properties of a passive, dispersive metamaterial that is matched to free space and has an index of refraction equal to zero. Thus, unlike previous zero-index investigations, both the permittivity and permeability are zero here at a specified frequency. One-, two-, and three-dimensional source problems are treated analytically. The one- and two-dimensional source problem results are confirmed numerically with finite difference time domain (FDTD) simulations. The FDTD simulator is also used to treat the corresponding one- and two-dimensional scattering problems. It is shown that in both the source and scattering configurations the electromagnetic fields in a matched zero-index medium take on a static character in space, yet remain dynamic in time, in such a manner that the underlying physics remains associated with propagating fields. Zero phase variation at various points in the zero-index medium is demonstrated once steady-state conditions are obtained. These behaviors are used to illustrate why a zero-index metamaterial, such as a zero-index electromagnetic band-gap structured medium, significantly narrows the far-field pattern associated with an antenna located within it. They are also used to show how a matched zero-index slab could be used to transform curved wave fronts into planar ones.
TL;DR: A phase-field model that can accurately simulate microstructural pattern formation for low-speed directional solidification of a dilute binary alloy is presented, and the addition of a phenomenological "antitrapping" solute current in the mass conservation relation is achieved.
Abstract: We present a detailed derivation and thin interface analysis of a phase-field model that can accurately simulate microstructural pattern formation for low-speed directional solidification of a dilute binary alloy. This advance with respect to previous phase-field models is achieved by the addition of a phenomenological "antitrapping" solute current in the mass conservation relation [Phys. Rev. Lett. 87, 115701 (2001)]]. This antitrapping current counterbalances the physical, albeit artificially large, solute trapping effect generated when a mesoscopic interface thickness is used to simulate the interface evolution on experimental length and time scales. Furthermore, it provides additional freedom in the model to suppress other spurious effects that scale with this thickness when the diffusivity is unequal in solid and liquid [SIAM J. Appl. Math. 59, 2086 (1999)]], which include surface diffusion and a curvature correction to the Stefan condition. This freedom can also be exploited to make the kinetic undercooling of the interface arbitrarily small even for mesoscopic values of both the interface thickness and the phase-field relaxation time, as for the solidification of pure melts [Phys. Rev. E 53, R3017 (1996)]]. The performance of the model is demonstrated by calculating accurately within a phase-field approach the Mullins-Sekerka stability spectrum of a planar interface and nonlinear cellular shapes for realistic alloy parameters and growth conditions.
TL;DR: It is shown that the recently proposed conceptually simple and easily calculated measure of permutation entropy can be effectively used to detect qualitative and quantitative dynamical changes.
Abstract: Timely detection of unusual and/or unexpected events in natural and man-made systems has deep scientific and practical relevance. We show that the recently proposed conceptually simple and easily calculated measure of permutation entropy can be effectively used to detect qualitative and quantitative dynamical changes. We illustrate our results on two model systems as well as on clinically characterized brain wave data from epileptic patients.
TL;DR: A very simple one-dimensional swimmer consisting of three spheres that are linked by rigid rods whose lengths can change between two values can be used in constructing molecular-sized machines.
Abstract: We propose a very simple one-dimensional swimmer consisting of three spheres that are linked by rigid rods whose lengths can change between two values. With a periodic motion in a nonreciprocal fashion, which breaks the time-reversal symmetry as well as the translational symmetry, we show that the model device can swim at low Reynolds number. This model system could be used in constructing molecular-sized machines.
Journal Article•
TL;DR: Spatial- and time-domain versions of the unidirectional pulse propagation equation (UPPE) are derived and compared from the point of view of their practical application in simulations of nonlinear optical pulse dynamics.
Abstract: Spatial- and time-domain versions of the unidirectional pulse propagation equation (UPPE) are derived and compared from the point of view of their practical application in simulations of nonlinear optical pulse dynamics. A modification of the UPPE suitable for ultrathin optical waveguides, such as submicron silica wires, is also presented. We show in detail how various, previously published propagation equations follow from the UPPE in a unified way that clearly elucidates their underlying approximations and areas of applicability.
TL;DR: The coupled dynamics of the internal states of a set of interacting elements and the network of interactions among them and the formation of a hierarchical interaction network that sustains a highly cooperative stationary state are explored.
Abstract: We explore the coupled dynamics of the internal states of a set of interacting elements and the network of interactions among them. Interactions are modeled by a spatial game and the network of interaction links evolves adapting to the outcome of the game. As an example, we consider a model of cooperation in which the adaptation is shown to facilitate the formation of a hierarchical interaction network that sustains a highly cooperative stationary state. The resulting network has the characteristics of a small world network when a mechanism of local neighbor selection is introduced in the adaptive network dynamics. The highly connected nodes in the hierarchical structure of the network play a leading role in the stability of the network. Perturbations acting on the state of these special nodes trigger global avalanches leading to complete network reorganization.
TL;DR: Through the study of airport network of China (ANC), composed of 128 airports and 1165 flights, it is shown the topological structure of ANC conveys two characteristics of small worlds, a short average path length and a high degree of clustering.
Abstract: Through the study of airport network of China (ANC), composed of 128 airports (nodes) and 1165 flights (edges), we show the topological structure of ANC conveys two characteristics of small worlds, a short average path length (2.067) and a high degree of clustering (0.733). The cumulative degree distributions of both directed and undirected ANC obey two-regime power laws with different exponents, i.e., the so-called double Pareto law. In-degrees and out-degrees of each airport have positive correlations, whereas the undirected degrees of adjacent airports have significant linear anticorrelations. It is demonstrated both weekly and daily cumulative distributions of flight weights (frequencies) of ANC have power-law tails. Besides, the weight of any given flight is proportional to the degrees of both airports at the two ends of that flight. It is also shown the diameter of each subcluster (consisting of an airport and all those airports to which it is linked) is inversely proportional to its density of connectivity. Efficiency of ANC and of its subclusters is measured through a simple definition. In terms of that, the efficiency of ANC's subclusters increases as the density of connectivity does. ANC is found to have an efficiency of 0.484.
TL;DR: It is found that the total contact area rises linearly with the load at small loads and the mean pressure in the contact regions is independent of load and proportional to the root-mean-square slope of the surface.
Abstract: Finite-element methods are used to study nonadhesive, frictionless contact between elastic solids with self-affine surfaces. We find that the total contact area rises linearly with the load at small loads. The mean pressure in the contact regions is independent of load and proportional to the root-mean-square slope of the surface. The constant of proportionality is nearly independent of the Poisson ratio and roughness exponent and lies between previous analytic predictions. The contact morphology is also analyzed. Connected contact regions have a fractal area and perimeter. The probability of finding a cluster of area a(c) drops as a(-tau )(c ) where tau increases with a decrease in roughness exponent. The distribution of pressures shows an exponential tail that is also found in many jammed systems. These results are contrasted to simpler models and experiments.
TL;DR: The simple pattern of the inclusion, the wide left-handed frequency band exhibited, and the low losses measured indicate the superiority of this inclusion in the realization of left- handed metamaterials.
Abstract: We analyze an $\mathsf{S}$-shaped inclusion for the realization of metamaterials exhibiting left-handed properties. Unlike most of the conventional inclusions used so far that are composed of two separate geometries---typically a split ring and a rod---the inclusion proposed in this paper is made of only one $\mathsf{S}$-shaped element which yields an overlapping negative permittivity and negative permeability response over a frequency band of about $2.6\phantom{\rule{0.3em}{0ex}}\mathrm{GHz}$. By adopting this geometry, we manage to lower the negative permittivity frequency band down to the level of the negative permeability frequency band, thus allowing the overlapping to occur. Therefore, the structure works as a stand alone and does not require the use of an additional rod. A theoretical analysis is carried out to study this inclusion and numerical simulations, as well as a Snell refraction experiment, clearly show that the material indeed exhibits a negative index of refraction at some frequencies. The simple pattern of the inclusion, the wide left-handed frequency band exhibited, and the low losses measured indicate the superiority of this inclusion in the realization of left-handed metamaterials.
TL;DR: A metric to quantify correlations between earthquakes is proposed and the original Omori law with p=1 emerges as a robust feature of seismicity, holding up to years even for aftershock sequences initiated by intermediate magnitude events.
Abstract: We propose a metric to quantify correlations between earthquakes. The metric consists of a product involving the time interval and spatial distance between two events, as well as the magnitude of the first one. According to this metric, events typically are strongly correlated to only one or a few preceding ones. Thus a classification of events as foreshocks, main shocks, or aftershocks emerges automatically without imposing predetermined space-time windows. In the simplest network construction, each earthquake receives an incoming link from its most correlated predecessor. The number of aftershocks for any event, identified by its outgoing links, is found to be scale free with exponent $\ensuremath{\gamma}=2.0(1)$. The original Omori law with $p=1$ emerges as a robust feature of seismicity, holding up to years even for aftershock sequences initiated by intermediate magnitude events. The broad distribution of distances between earthquakes and their linked aftershocks suggests that aftershock collection with fixed space windows is not appropriate.
TL;DR: The bottleneck effect is shown to be equally strong both for magnetic and nonmagnetic turbulence, but it is far weaker in one-dimensional spectra that are normally studied in laboratory turbulence.
Abstract: Nonhelical hydromagnetic forced turbulence is investigated using large scale simulations on up to $256$ processors and ${1024}^{3}$ mesh points. The magnetic Prandtl number is varied between 1∕8 and 30, although in most cases it is unity. When the magnetic Reynolds number is based on the inverse forcing wave number, the critical value for dynamo action is shown to be around 35 for magnetic Prandtl number of unity. For small magnetic Prandtl numbers we find the critical magnetic Reynolds number to increase with decreasing magnetic Prandtl number. The Kazantsev ${k}^{3∕2}$ spectrum for magnetic energy is confirmed for the kinematic regime, i.e., when nonlinear effects are still unimportant and when the magnetic Prandtl number is unity. In the nonlinear regime, the energy budget converges for large Reynolds numbers (around 1000) such that for our parameters about $70%$ is in kinetic energy and about $30%$ is in magnetic energy. The energy dissipation rates are converged to $30%$ viscous dissipation and $70%$ resistive dissipation. Second-order structure functions of the Elsasser variables give evidence for a ${k}^{\ensuremath{-}5∕3}$ spectrum. Nevertheless, the three-dimensional spectrum is close to ${k}^{\ensuremath{-}3∕2}$, but we argue that this is due to the bottleneck effect. The bottleneck effect is shown to be equally strong both for magnetic and nonmagnetic turbulence, but it is far weaker in one-dimensional spectra that are normally studied in laboratory turbulence. Structure function exponents for other orders are well described by the She-Leveque formula, but the velocity field is significantly less intermittent and the magnetic field is more intermittent than the Elsasser variables.
TL;DR: From the simulation of space-time evolution of the vehicle headway, it is shown that the traffic jam is suppressed efficiently with taking into account the information about the motion of more vehicles in front, and the analytical result is consonant with the simulation one.
Abstract: An extended car following model is proposed by incorporating an intelligent transportation system in traffic. The stability condition of this model is obtained by using the linear stability theory. The results show that anticipating the behavior of more vehicles ahead leads to the stabilization of traffic systems. The modified Korteweg-de Vries equation (the mKdV equation, for short) near the critical point is derived by applying the reductive perturbation method. The traffic jam could be thus described by the kink-antikink soliton solution for the mKdV equation. From the simulation of space-time evolution of the vehicle headway, it is shown that the traffic jam is suppressed efficiently with taking into account the information about the motion of more vehicles in front, and the analytical result is consonant with the simulation one.