Randomized algorithm to determine the eigenvector of a stochastic matrix with application to the PageRank problem

Consideration was given to estimation of the eigenvector corresponding to the greatest eigenvalue of a stochastic matrix. There exist numerous applications of this problem arising at ranking the results of search, coordination of the multiagent system actions, network control, and data analysis. The standard technique for its solution comes to the power method with an additional regularization of the original matrix. A new randomized algorithm was proposed, and a uniform—over the entire class of the stochastic matrices of a given size—upper boundary of the convergence rate was validated. It is given by {ie342-1}, where C is an absolute constant, N is size, and n is the number of iterations. This boundary seems promising because ln(N) is smallish even for a very great size. The algorithm relies on the mirror descent method for the problems of convex stochastic optimization. Applicability of the method to the PageRank problem of ranking the Internet pages was discussed.     

A parametric filtering algorithm for the graph isomorphism problem

We introduce a new filtering algorithm, called IDL(d)-filtering, for a global constraint dedicated to the graph isomorphism problem—the goal of which is to decide if two given graphs have an identical structure. The basic idea of IDL(d)-filtering is to label every vertex with respect to its relationships with other vertices around it in the graph, and to use these labels to filter domains by removing values that have different labels. IDL(d)-filtering is parameterized by a positive integer value d which gives a limit on the distance between a vertex to be labelled and the set of vertices considered to build its label. We experimentally compare different instantiations of IDL(d)-filtering with state-of-the-art dedicated algorithms and show that IDL(d)-filtering is more efficient on regular sparse graphs and competitive on other kinds of graphs.     

Nested clusters with intercluster routing

Very large scale networks have become common in distributed systems. To efficiently manage these networks, various techniques are being developed in the distributed and networking research community. In this paper, we focus on one of those techniques, network clustering, i.e., the partitioning of a system into connected subsystems. The clustering we compute is size-oriented: given a parameter K of the algorithm, we compute, as far as possible, clusters of size K. We present an algorithm to compute a binary hierarchy of nested disjoint clusters. A token browses the network and recruits nodes to its cluster. When a cluster reaches a maximal size defined by a parameter of the algorithm, it is divided when possible, and tokens are created in both of the new clusters. The new clusters are then built and divided in the same fashion. The token browsing scheme chosen is a random walk, in order to ensure local load balancing. To allow the division of clusters, a spanning tree is built for each cluster. At each division, information on how to route messages between the clusters is stored. The naming process used for the clusters, along with the information stored during each division, allows routing between any two clusters.     

C-Rank:一种Deep Web数据记录可信度评估方法

针对Web信息可信度问题,提出了一种为Deep Web数据记录计算可信度的有效方法C-Rank。该方法为每一条记录构造一个S-R可信度网络,包含两种类型顶点及三种类型边。首先基于可信度传播的思想,利用顶点出度为每一个顶点计算其局部可信度值;再利用Record顶点入度及相邻Site顶点的可信度值,为该Record顶点计算权值;继而求得整个S-R网络的全局可信度值。实验证明,C-Rank方法能够合理而有效地评价数据记录的可信度,从而达到甄别虚假信息,为用户推荐可信数据记录的目的。该方法普遍适用于Deep Web的各个领域。     

Streaming multiple aggregations using phantoms

Data streams characterize the high speed and large volume input of a new class of applications such as network monitoring, web content analysis and sensor networks. Among these applications, network monitoring may be the most compelling one—the backbone of a large internet service provider can generate 1 petabyte of data per day. For many network monitoring tasks such as traffic analysis and statistics collection, aggregation is a primitive operation. Various analytical and statistical needs naturally lead to related aggregate queries. In this article, we address the problem of efficiently computing multiple aggregations over high-speed data streams based on the two-level query processing architecture of GS, a real data stream management system deployed in AT & T. We discern that additionally computing and maintaining fine-granularity aggregations (called phantoms) has the benefit of supporting shared computation. Based on a thorough analysis, we propose algorithms to identify the best set of phantoms to maintain and determine allocation of resources (particularly, space) to compute the aggregations. Experiments show that our algorithm achieves near-optimal computation costs, which outperforms the best adapted algorithm by more than an order of magnitude.     

Approximation Algorithms for Degree-Constrained Minimum-Cost Network-Design Problems

We study network-design problems with two different design objectives: the total cost of the edges and nodes in the network and the maximum degree of any node in the network. A prototypical example is the degree-constrained node-weighted Steiner tree problem: We are given an undirected graph G(V,E) , with a nonnegative integral function d that specifies an upper bound d(v) on the degree of each vertex v ∈ V in the Steiner tree to be constructed, nonnegative costs on the nodes, and a subset of k nodes called terminals. The goal is to construct a Steiner tree T containing all the terminals such that the degree of any node v in T is at most the specified upper bound d(v) and the total cost of the nodes in T is minimum. Our main result is a bicriteria approximation algorithm whose output is approximate in terms of both the degree and cost criteria—the degree of any node v ∈ V in the output Steiner tree is O(d(v) log k) and the cost of the tree is O(log k) times that of a minimum-cost Steiner tree that obeys the degree bound d(v) for each node v . Our result extends to the more general problem of constructing one-connected networks such as generalized Steiner forests. We also consider the special case in which the edge costs obey the triangle inequality and present simple approximation algorithms with better performance guarantees. Received December 21, 1998; revised September 24, 1999.     

Periodic Merging Networks

We consider the problem of merging two sorted sequences on constant degree networks performing compare—exchange operations only. The classical solution to this problem is given by the networks based on Batcher's Odd—Even Merge and Bitonic Merge running in log(2n ) time. Due to the obvious log n lower bound for the runtime, this is time-optimal. We present a new family of merging networks working in a completely different way than the previously known algorithms. They have a novel property of being periodic: this means that for some (small) constant k , each processing unit of the network performs the same operations at steps t and t+k (as long as t+k does not exceed the runtime). The only operations executed are compare—exchange operations, just like in the case of Batcher's networks. The architecture of the networks is very simple and easy to lay out. We show that even for period 3 there is a network in our family merging two n -element sorted sequences in time O(log n ). Since each network of period 2 requires steps to merge such sequences, 3 is the smallest period for which we may achieve a fast runtime. In order to improve constants standing in front of log n we increment the period and tune the construction using additional techniques. We achieve the runtime 9 . . . log_3 n 5.7 . . . log n for a network of period 4, and 2.25 . . . ((k+3)/(k-1+log 3))log n 2.25 . . . log n for a network of period k+3 , for . Due to the periodic design, our networks have small area complexity. For instance, if each processing unit requires O(1) area and a comparator uses a single wire of width O(1) connecting the processing elements, then our networks require area. This compares well with Batcher's networks that require area . Received February 1997, and in revised form September 1997, and in final form February 1998.     

Supervised clustering of label ranking data using label preference information

This paper studies supervised clustering in the context of label ranking data. The goal is to partition the feature space into K clusters, such that they are compact in both the feature and label ranking space. This type of clustering has many potential applications. For example, in target marketing we might want to come up with K different offers or marketing strategies for our target audience. Thus, we aim at clustering the customers’ feature space into K clusters by leveraging the revealed or stated, potentially incomplete customer preferences over products, such that the preferences of customers within one cluster are more similar to each other than to those of customers in other clusters. We establish several baseline algorithms and propose two principled algorithms for supervised clustering. In the first baseline, the clusters are created in an unsupervised manner, followed by assigning a representative label ranking to each cluster. In the second baseline, the label ranking space is clustered first, followed by partitioning the feature space based on the central rankings. In the third baseline, clustering is applied on a new feature space consisting of both features and label rankings, followed by mapping back to the original feature and ranking space. The RankTree principled approach is based on a Ranking Tree algorithm previously proposed for label ranking prediction. Our modification starts with K random label rankings and iteratively splits the feature space to minimize the ranking loss, followed by re-calculation of the K rankings based on cluster assignments. The MM-PL approach is a multi-prototype supervised clustering algorithm based on the Plackett-Luce (PL) probabilistic ranking model. It represents each cluster with a union of Voronoi cells that are defined by a set of prototypes, and assign each cluster with a set of PL label scores that determine the cluster central ranking. Cluster membership and ranking prediction for a new instance are determined by cluster membership of its nearest prototype. The unknown cluster PL parameters and prototype positions are learned by minimizing the ranking loss, based on two variants of the expectation-maximization algorithm. Evaluation of the proposed algorithms was conducted on synthetic and real-life label ranking data by considering several measures of cluster goodness: (1) cluster compactness in feature space, (2) cluster compactness in label ranking space and (3) label ranking prediction loss. Experimental results demonstrate that the proposed MM-PL and RankTree models are superior to the baseline models. Further, MM-PL is has shown to be much better than other algorithms at handling situations with significant fraction of missing label preferences.     

Mixing local and global information for community detection in large networks

Clustering networks play a key role in many scientific fields, from Biology to Sociology and Computer Science. Some clustering approaches are called global because they exploit knowledge about the whole network topology. Vice versa, so-called local methods require only a partial knowledge of the network topology. Global approaches yield accurate results but do not scale well on large networks; local approaches, vice versa, are less accurate but computationally fast. We propose CONCLUDE (COmplex Network CLUster DEtection), a new clustering method that couples the accuracy of global approaches with the scalability of local methods. CONCLUDE generates random, non-backtracking walks of finite length to compute the importance of each edge in keeping the network connected, i.e., its edge centrality. Edge centralities allow for mapping vertices onto points of a Euclidean space and compute all-pairs distances between vertices; those distances are then used to partition the network into clusters.     

On the Planarization of Wireless Sensor Networks

Network planarization has been an important technique in numerous sensornet protocols—such as Greedy Perimeter Stateless Routing (GPSR), topology discovery, data-centric storage, etc.—however the planarization process itself has been difficult. Known efficient planarization algorithms exist only for restrictive wireless network models: unit-disk graphs with accurately known location information. In this paper, we study efficient planarization of wireless sensor networks, and present a novel planarization method for a more general network model, where sensors can have non-uniform transmission ranges and no location information is needed. Our planarization algorithms also include a (2+ε)-approximation algorithm and an FPT algorithm for the bipartite planarization problem.     

A branch‐and‐price approach to k‐clustering minimum biclique completion problem

We consider the problem of finding k‐bipartite subgraphs, called “clusters,” in a bipartite graph , such that each vertex i of S appears in exactly one of the subgraphs, every vertex j of T appears in each cluster in which at least one of its neighbors appears, and the total number of edges needed to complete each cluster (i.e. to become a biclique) is minimized. This problem has been shown to be strongly NP‐hard and is known as the k‐clustering minimum biclique completion problem. It has been applied to bundling channels for multicast transmissions. The application consists of finding k‐multicast sessions that partition the set of demands, given a set of demands of services from clients. Each service should belong to a single multicast session, while each client can appear in more than one session. We extend previous work by developing a branch‐and‐price algorithm that embeds a new metaheuristic based on variable neighborhood infeasible search and a branching rule that exploits the problem structure. The metaheuristic can also efficiently solve the pricing subproblem. In addition to the random instances used in the literature, we present structured instances generated using the MovieLens dataset collected by the GroupLens Research Project. Extensive computational results show that our branch‐and‐price algorithm outperforms the approaches proposed in the literature.     

Coloring unstructured radio networks

During and immediately after their deployment, ad hoc and sensor networks lack an efficient communication scheme rendering even the most basic network coordination problems difficult. Before any reasonable communication can take place, nodes must come up with an initial structure that can serve as a foundation for more sophisticated algorithms. In this paper, we consider the problem of obtaining a vertex coloring as such an initial structure. We propose an algorithm that works in the unstructured radio network model. This model captures the characteristics of newly deployed ad hoc and sensor networks, i.e. asynchronous wake-up, no collision-detection, and scarce knowledge about the network topology. When modeling the network as a graph with bounded independence, our algorithm produces a correct coloring with O(Δ) colors in time O(Δ log n) with high probability, where n and Δ are the number of nodes in the network and the maximum degree, respectively. Also, the number of locally used colors depends only on the local node density. Graphs with bounded independence generalize unit disk graphs as well as many other well-known models for wireless multi-hop networks. They allow us to capture aspects such as obstacles, fading, or irregular signal-propagation. A preliminary version of this work has been published in [20] as Coloring Unstructured Radio Networks, In Proceedings of the 17th Symposium on Parallel Algorithms and Architectures (SPAA), Las Vegas, Nevada, 2005.     

Fast parallel algorithms for graph similarity and matching

This paper addresses the problem of global graph alignment on supercomputer-class clusters. We define the alignment of two graphs, as a mapping of each vertex in the first graph to a unique vertex in the second graph so as to optimize a given similarity-based cost function.¹ Using a state of the art serial algorithm for the computation of vertex similarity scores called Network Similarity Decomposition (NSD), we derive corresponding parallel formulations. Coupling this parallel similarity algorithm with a parallel auction-based bipartite matching technique, we obtain a highly efficient and scalable graph matching pipeline. We validate the performance of our integrated approach on a large parallel platform and on diverse graph instances (including Protein Interaction, Wikipedia and Web networks). Experimental results demonstrate that our algorithms scale to large machine configurations (thousands of cores) and problem instances, enabling the alignment of networks of sizes two orders of magnitude larger than reported in the current literature.     

On positive influence dominating sets in social networks

In this paper, we investigate the positive influence dominating set (PIDS) which has applications in social networks. We prove that PIDS is APX-hard and propose a greedy algorithm with an approximation ratio of H(δ) where H is the harmonic function and δ is the maximum vertex degree of the graph representing a social network.     

The mailbox problem

We propose and solve a synchronization problem called the mailbox problem, motivated by a particular type of interaction between a processor and an external device or between two threads. In this problem, a postman delivers letters to the mailbox of a home owner and uses a flag to signal a non-empty mailbox. The owner must remove all letters delivered to the mailbox and should not walk to the mailbox if it is empty. We present algorithms and an impossibility result for this problem.     

一种加权稠密子图社区发现算法

目前,针对复杂网络的社区发现算法大多仅根据网络的拓扑结构来确定社区,然而现实复杂网络中的边可能带有表示连接紧密程度或者可信度意义的权重,这些先验信息对社区发现的准确性至关重要.针对该问题,提出了基于加权稠密子图的重叠聚类算法（overlap community detection on weighted networks,简称OCDW）.首先,综合考虑网络拓扑结构及真实网络中边权重的影响,给出了一种网络中边的权重定义方法;进而给出种子节点选取方式和权重更新策略;最终得到聚类结果.OCDW算法在无权网络和加权网络都适用.通过与一些经典的社区发现算法在9个真实网络数据集上进行分析比较,结果表明算法OCDW在F度量、准确度、分离度、标准互信息、调整兰德系数、模块性及运行时间等方面均表现出较好的性能.     

Geno-mathematical identification of the multi-layer perceptron

In this paper, we will focus on the use of the three-layer backpropagation network in vector-valued time series estimation problems. The neural network provides a framework for noncomplex calculations to solve the estimation problem, yet the search for optimal or even feasible neural networks for stochastic processes is both time consuming and uncertain. The backpropagation algorithm—written in strict ANSI C—has been implemented as a standalone support library for the genetic hybrid algorithm (GHA) running on any sequential or parallel main frame computer. In order to cope with ill-conditioned time series problems, we extended the original backpropagation algorithm to a K nearest neighbors algorithm (K-NARX), where the number K is determined genetically along with a set of key parameters. In the K-NARX algorithm, the terminal solution at instant t can be used as a starting point for the next t, which tends to stabilize the optimization process when dealing with autocorrelated time series vectors. This possibility has proved to be especially useful in difficult time series problems. Following the prevailing research directions, we use a genetic algorithm to determine optimal parameterizations for the network, including the lag structure for the nonlinear vector time series system, the net structure with one or two hidden layers and the corresponding number of nodes, type of activation function (currently the standard logistic sigmoid, a bipolar transformation, the hyperbolic tangent, an exponential function and the sine function), the type of minimization algorithm, the number K of nearest neighbors in the K-NARX procedure, the initial value of the Levenberg–Marquardt damping parameter and the value of the neural learning (stabilization) coefficient α. We have focused on a flexible structure allowing addition of, e.g., new minimization algorithms and activation functions in the future. We demonstrate the power of the genetically trimmed K-NARX algorithm on a representative data set.     

Vertex rankings of chordal graphs and weighted trees

In this paper we consider the vertex ranking problem of weighted trees. We show that this problem is strongly NP-hard. We also give a polynomial-time reduction from the problem of vertex ranking of weighted trees to the vertex ranking of (simple) chordal graphs, which proves that the latter problem is NP-hard. In this way we solve an open problem of Aspvall and Heggernes. We use this reduction and the algorithm of Bodlaender et al.'s for vertex ranking of partial k-trees to give an exact polynomial-time algorithm for vertex ranking of a tree with bounded and integer valued weight functions. This algorithm serves as a procedure in designing a PTAS for weighted vertex ranking problem of trees with bounded weight functions.