A multilevel algorithm for force-directed graph-drawing

We describe a heuristic method for drawing graphs which uses a multilevel framework combined with a force-directed placement algorithm. The multilevel technique matches and coalesces pairs of adjacent vertices to define a new graph and is repeated recursively to create a hierarchy of increasingly coarse graphs, G0, G1, …, GL. The coarsest graph, GL, is then given an initial layout and the layout is refined and extended to all the graphs starting with the coarsest and ending with the original. At each successive change of level, l, the initial layout for Gl is taken from its coarser and smaller child graph, Gl+1, and refined using force-directed placement. In this way the multilevel framework both accelerates and appears to give a more global quality to the drawing. The algorithm can compute both 2 & 3 dimensional layouts and we demonstrate it on examples ranging in size from 10 to 225,000 vertices. It is also very fast and can compute a 2D layout of a sparse graph in around 12 seconds for a 10,000 vertex graph to around 5-7 minutes for the largest graphs. This is an order of magnitude faster than recent implementations of force-directed placement algorithms.

A combined evolutionary search and multilevel optimisation approach to graph-partitioning

The graph-partitioning problem is to divide a graph into several pieces so that the number of vertices in each piece is the same within some defined tolerance and the number of cut edges is minimised. Important applications of the problem arise, for example, in parallel processing where data sets need to be distributed across the memory of a parallel machine. Very effective heuristic algorithms have been developed for this problem which run in real-time, but it is not known how good the partitions are since the problem is, in general, NP-complete. This paper reports an evolutionary search algorithm for finding benchmark partitions. A distinctive feature is the use of a multilevel heuristic algorithm to provide an effective crossover. The technique is tested on several example graphs and it is demonstrated that our method can achieve extremely high quality partitions significantly better than those found by the state-of-the-art graph-partitioning packages.

Hall conditions for edge-weighted bipartite graphs

A weighted variant of Hall's condition for the existence of matchings is shown to be equivalent to the existence of a matching in a lexicographic product. This is used to introduce characterizations of those bipartite graphs whose edges may be replicated so as to yield semiregular multigraphs or, equivalently, semiregular edge-weightings. Such bipartite graphs will be called semiregularizable. Some infinite families of semiregularizable trees are described and all semiregularizable trees on at most 11 vertices are listed. Matrix analogues of some of the results are mentioned and are shown to imply some of the known characterizations of regularizable graphs.

Vibrational properties of nanometric AB(2) ionic clusters

A broad survey of harmonic dynamics in AB(2) clusters with up to N = 3000 atoms is performed using a simple rigid ion model, with ionic radii selected to give rutile as the ground state structure for the corresponding extended crystal. The vibrational density of states is already close to its bulk counterpart for N similar to 500, with characteristic differences due to surfaces, edges and vertices. Two methods are proposed and tested to map the cluster vibrational states onto the rutile crystal phonons. The net distinction between infrared (IR) active and Raman active modes that exists for bulk rutile becomes more and more blurred as the cluster size is reduced. It is found that, in general, the higher the IR activity of the mode, the more this is affected by the system size. IR active modes are found to spread over a wide frequency range for the finite clusters. Simple models based on either a crude confinement constraint or surface pressure arguments fail to reproduce the results of the calculations. The effects of the stoichiometry and dielectric properties of the surrounding medium on the vibrational properties of the clusters are also investigated.

Formation of NH4[Hg-3(NH)(2)](NO3)(3) and transformation to [Hg2N](NO3)

NH4[Hg-3(NH)(2)](NO3)(3) (1) and [Hg2N](NO3) (2) are obtained from cone. aqueous ammonia solutions of Hg(NO3)(2) at ambient temperature and under hydrothermal conditions at 180 degreesC, respectively, as colourless and dark yellow to light brown single crystals. The crystal structures {NH4[Hg-3(NH)(2)](NO3)(3): cubic, P4(I)32, a = 1030.4(2) pm, Z = 4, R-all = 0.028; [Hg2N](NO3): tetragonal, P4(3)2(1)2, a = 1540.4(1), c = 909.8(1) pm, Z = 4, R-all = 0.054} have been determined from single crystal data. Both exhibit network type structures in which [HNHg3] and [NHg4] tetrahedra of the partial structures of 1 and 2 are connected via three and four vertices, respectively. 1 transforms at about 270 degreesC in a straightforward reaction to 2 whereby the decomposition products of NH4NO3 are set free. 2 decomposes at about 380 degreesC forming yellow HgO. Most certainly, I is identical with a mineral previously analyzed as

Connections between Classical and Parametric Network Entropies

This paper explores relationships between classical and parametric measures of graph (or network) complexity. Classical measures are based on vertex decompositions induced by equivalence relations. Parametric measures, on the other hand, are constructed by using information functions to assign probabilities to the vertices. The inequalities established in this paper relating classical and parametric measures lay a foundation for systematic classification of entropy-based measures of graph complexity.

The Forgiving Graph: a distributed data structure for low stretch under adversarial attack

We consider the problem of self-healing in peer-to-peer networks that are under repeated attack by an omniscient adversary. We assume that, over a sequence of rounds, an adversary either inserts a node with arbitrary connections or deletes an arbitrary node from the network. The network responds to each such change by quick “repairs,” which consist of adding or deleting a small number of edges. These repairs essentially preserve closeness of nodes after adversarial deletions, without increasing node degrees by too much, in the following sense. At any point in the algorithm, nodes v and w whose distance would have been l in the graph formed by considering only the adversarial insertions (not the adversarial deletions), will be at distance at most l log n in the actual graph, where n is the total number of vertices seen so far. Similarly, at any point, a node v whose degree would have been d in the graph with adversarial insertions only, will have degree at most 3d in the actual graph. Our distributed data structure, which we call the Forgiving Graph, has low latency and bandwidth requirements. The Forgiving Graph improves on the Forgiving Tree distributed data structure from Hayes et al. (2008) in the following ways: 1) it ensures low stretch over all pairs of nodes, while the Forgiving Tree only ensures low diameter increase; 2) it handles both node insertions and deletions, while the Forgiving Tree only handles deletions; 3) it requires only a very simple and minimal initialization phase, while the Forgiving Tree initially requires construction of a spanning tree of the network.

The Forgiving Graph:A distributed data structure for low stretch under adversarial attack

We consider the problem of self-healing in peer-to-peer networks that are under repeated attack by an omniscient adversary. We assume that, over a sequence of rounds, an adversary either inserts a node with arbitrary connections or deletes an arbitrary node from the network. The network responds to each such change by quick "repairs," which consist of adding or deleting a small number of edges. These repairs essentially preserve closeness of nodes after adversarial deletions,without increasing node degrees by too much, in the following sense. At any point in the algorithm, nodes v and w whose distance would have been - in the graph formed by considering only the adversarial insertions (not the adversarial deletions), will be at distance at most - log n in the actual graph, where n is the total number of vertices seen so far. Similarly, at any point, a node v whose degreewould have been d in the graph with adversarial insertions only, will have degree at most 3d in the actual graph. Our distributed data structure, which we call the Forgiving Graph, has low latency and bandwidth requirements. The Forgiving Graph improves on the Forgiving Tree distributed data structure from Hayes et al. (2008) in the following ways: 1) it ensures low stretch over all pairs of nodes, while the Forgiving Tree only ensures low diameter increase; 2) it handles both node insertions and deletions, while the Forgiving Tree only handles deletions; 3) it requires only a very simple and minimal initialization phase, while the Forgiving Tree initially requires construction of a spanning tree of the network. © Springer-Verlag 2012.

Exploring Vertex Interactions in Ferroelectric Flux-Closure Domains

Using piezoresponse force microscopy, we have observed the progressive development of ferroelectric flux-closure domain structures and Landau−Kittel-type domain patterns, in 300 nm thick single-crystal BaTiO3 platelets. As the microstructural development proceeds, the rate of change of the domain configuration is seen to decrease exponentially. Nevertheless, domain wall velocities throughout are commensurate with creep processes in oxide ferroelectrics. Progressive screening of macroscopic destabilizing fields, primarily the surface-related depolarizing field, successfully describes the main features of the observed kinetics. Changes in the separation of domain-wall vertex junctions prompt a consideration that vertex−vertex interactions could be influencing the measured kinetics. However, the expected dynamic signatures associated with direct vertex−vertex interactions are not resolved. If present, our measurements confine the length scale for interaction between vertices to the order of a few hundred nanometers.

Finding the Most Descriptive Substructures in Graphs with Discrete and Numeric Labels

Many graph datasets are labelled with discrete and numeric attributes. Most frequent substructure discovery algorithms ignore numeric attributes; in this paper we show how they can be used to improve search performance and discrimination. Our thesis is that the most descriptive substructures are those which are normative both in terms of their structure and in terms of their numeric values. We explore the relationship between graph structure and the distribution of attribute values and propose an outlier-detection step, which is used as a constraint during substructure discovery. By pruning anomalous vertices and edges, more weight is given to the most descriptive substructures. Our method is applicable to multi-dimensional numeric attributes; we outline how it can be extended for high-dimensional data. We support our findings with experiments on transaction graphs and single large graphs from the domains of physical building security and digital forensics, measuring the effect on runtime, memory requirements and coverage of discovered patterns, relative to the unconstrained approach.

Vertex Clustering of Augmented Graph Streams

In this paper we propose a graph stream clustering algorithm with a unied similarity measure on both structural and attribute properties of vertices, with each attribute being treated as a vertex. Unlike others, our approach does not require an input parameter for the number of clusters, instead, it dynamically creates new sketch-based clusters and periodically merges existing similar clusters. Experiments on two publicly available datasets reveal the advantages of our approach in detecting vertex clusters in the graph stream. We provide a detailed investigation into how parameters affect the algorithm performance. We also provide a quantitative evaluation and comparison with a well-known offline community detection algorithm which shows that our streaming algorithm can achieve comparable or better average cluster purity.

Generating Realistic Labelled, Weighted Random Graphs

Generative algorithms for random graphs have yielded insights into the structure and evolution of real-world networks. Most networks exhibit a well-known set of properties, such as heavy-tailed degree distributions, clustering and community formation. Usually, random graph models consider only structural information, but many real-world networks also have labelled vertices and weighted edges. In this paper, we present a generative model for random graphs with discrete vertex labels and numeric edge weights. The weights are represented as a set of Beta Mixture Models (BMMs) with an arbitrary number of mixtures, which are learned from real-world networks. We propose a Bayesian Variational Inference (VI) approach, which yields an accurate estimation while keeping computation times tractable. We compare our approach to state-of-the-art random labelled graph generators and an earlier approach based on Gaussian Mixture Models (GMMs). Our results allow us to draw conclusions about the contribution of vertex labels and edge weights to graph structure.

Main eigenvalues and (κ, τ)-regular sets

A (κ, τ)-regular set is a subset of the vertices of a graph G, inducing a κ-regular subgraph such that every vertex not in the subset has τ neighbors in it. A main eigenvalue of the adjacency matrix A of a graph G has an eigenvector not orthogonal to the all-one vector j. For graphs with a (κ, τ)-regular set a necessary and sufficient condition for an eigenvalue be non-main is deduced and the main eigenvalues are characterized. These results are applied to the construction of infinite families of bidegreed graphs with two main eigenvalues and the same spectral radius (index) and some relations with strongly regular graphs are obtained. Finally, the determination of (κ, τ)-regular sets is analyzed. © 2009 Elsevier Inc. All rights reserved.

Politopo de Birkhoff acíclico

Neste trabalho estabelece-se uma interpreta c~ao geom etrica, em termos da teoria dos grafos, para v ertices, arestas e faces de uma qualquer dimens~ao do politopo de Birkho ac clico, Tn = n(T), onde T e uma arvore com n v ertices. Generaliza-se o resultado obtido por G. Dahl, [18], para o c alculo do di^ametro do grafo G( t n), onde t n e o politopo das matrizes tridiagonais duplamente estoc asticas. Adicionalmente, para q = 0; 1; 2; 3 s~ao obtidas f ormulas expl citas para a contagem do n umero de q faces do politopo de Birkho tridiagonal, t n, e e feito o estudo da natureza geom etrica dessas mesmas faces. S~ao, tamb em, apresentados algoritmos para efectuar contagens do n umero de faces de dimens~ao inferior a de uma dada face do politopo de Birkho ac clico.

Majorantes para a ordem de subgrafos induzidos k-regulares

Muitos dos problemas de otimização em grafos reduzem-se à determinação de um subconjunto de vértices de cardinalidade máxima que induza um subgrafo k-regular. Uma vez que a determinação da ordem de um subgrafo induzido k-regular de maior ordem é, em geral, um problema NP-difícil, são deduzidos novos majorantes, a determinar em tempo polinomial, que em muitos casos constituam boas aproximações das respetivas soluções ótimas. Introduzem-se majorantes espetrais usando uma abordagem baseada em técnicas de programação convexa e estabelecem-se condições necessárias e suficientes para que sejam atingidos. Adicionalmente, introduzem-se majorantes baseados no espetro das matrizes de adjacência, laplaciana e laplaciana sem sinal. É ainda apresentado um algoritmo não polinomial para a determinação de umsubconjunto de vértices de umgrafo que induz umsubgrafo k-regular de ordem máxima para uma classe particular de grafos. Finalmente, faz-se um estudo computacional comparativo com vários majorantes e apresentam-se algumas conclusões.