Non-adjacent Nodes Research Articles

A hybrid method of link prediction in directed graphs

Link prediction is an important issue in complex network analysis and mining. Given the structure of a network, a link prediction algorithm obtains the probability that a link is established between two non-adjacent nodes in the future snapshots of the network. Many of the available link prediction methods are based on common neighborhood. A problem with these methods is that if two nodes do not have any common neighbors, they always predict a chance of zero for establishment of a link between them; however, such nodes have been shown to establish links in some real systems. Another issue with these measures is that they often disregard the connection direction. Here, we propose a novel measure based on common neighborhood that resolves the above issues. The proposed measures are applied on three benchmark networks in both unsupervised and supervised learning modes. Our experiments show the superior performance of the proposed measures over that of the state-of-the-art link prediction methods.

A tractable latent variable model for nonlinear dimensionality reduction

We propose a latent variable model to discover faithful low-dimensional representations of high-dimensional data. The model computes a low-dimensional embedding that aims to preserve neighborhood relationships encoded by a sparse graph. The model both leverages and extends current leading approaches to this problem. Like t-distributed Stochastic Neighborhood Embedding, the model can produce two- and three-dimensional embeddings for visualization, but it can also learn higher-dimensional embeddings for other uses. Like LargeVis and Uniform Manifold Approximation and Projection, the model produces embeddings by balancing two goals-pulling nearby examples closer together and pushing distant examples further apart. Unlike these approaches, however, the latent variables in our model provide additional structure that can be exploited for learning. We derive an Expectation-Maximization procedure with closed-form updates that monotonically improve the model's likelihood: In this procedure, embeddings are iteratively adapted by solving sparse, diagonally dominant systems of linear equations that arise from a discrete graph Laplacian. For large problems, we also develop an approximate coarse-graining procedure that avoids the need for negative sampling of nonadjacent nodes in the graph. We demonstrate the model's effectiveness on datasets of images and text.

A Scalable Similarity-Popularity Link Prediction Method

Link prediction is the task of computing the likelihood that a link exists between two given nodes in a network. With countless applications in different areas of science and engineering, link prediction has received the attention of many researchers working in various disciplines. Considerable research efforts have been invested into the development of increasingly accurate prediction methods. Most of the proposed algorithms, however, have limited use in practice because of their high computational requirements. The aim of this work is to develop a scalable link prediction algorithm that offers a higher overall predictive power than existing methods. The proposed solution falls into the class of global, parameter-free similarity-popularity-based methods, and in it, we assume that network topology is governed by three factors: popularity of the nodes, their similarity and the attraction induced by local neighbourhood. In our approach, popularity and neighbourhood-caused attraction are computed directly from the network topology and factored out by introducing a specific weight map, which is then used to estimate the dissimilarity between non-adjacent nodes through shortest path distances. We show through extensive experimental testing that the proposed method produces highly accurate predictions at a fraction of the computational cost required by existing global methods and at a low additional cost compared to local methods. The scalability of the proposed algorithm is demonstrated on several large networks having hundreds of thousands of nodes.

Meta-Learning on Graph with Curvature-Based Analysis (Student Abstract)

Learning latent representations in graphs is finding a mapping that embeds nodes or edges as data points in a low-dimensional vector space. This paper introduces a flexible framework to enhance existing methodologies that have difficulty capturing local proximity and global relationships at the same time. Our approach generates a virtual edge between non-adjacent nodes based on the Forman-Ricci curvature in network. By analyzing the network using topological information, global relationships structurally similar can easily be detected and successfully integrated with previous works.

Analysis on the High-order Distributed Consensus under Probabilistic Quantized Communication

Because the high-order distributed consensus algorithm utilizes the previous state values of non-adjacent nodes to accelerate the convergent rate, under the quantized communication, it means that the high-order distributed consensus algorithm utilizes the quantized information of more non-adjacent nodes, the performance of the high-order algorithm become very important under quantized communication. in the paper, the performance and the mean square errors of the high-order probabilistic quantization distributed consensus algorithm were analyzed. By analysis and simulation, the bigger the length of quantization message is, the better the performance is, and the smaller the square errors of the high-order algorithm is.

Diffusion LMS algorithms with multi combination for distributed estimation: Formulation and performance analysis

We propose diffusion least-mean-square (LMS) algorithms that use multi-combination step. We allow each node in the network to use information from multi-hop neighbors to approximate a global cost function accurately. By minimizing this cost and dividing multi-hop range summation into 1-hop range combination steps, we derive new diffusion LMS algorithms. The resulting distributed algorithms consist of adaptation and multi-combination step. Multi combination allows each node to use information from non-adjacent nodes at each time instant, thereby reducing steady-state error. We analyzed the output to derive stability conditions and to quantify the transient and steady-state behaviors. Theoretical and experimental results indicate that the proposed algorithms have lower steady-state error compared to the conventional diffusion LMS algorithms. We also propose a new combination rule for the multi-combination step which can further improve the estimation performance of the proposed algorithms.

New relationships for multi-neighborhood search for the minimum linear arrangement problem

We develop a series of theorems about the graph structure of the classical Minimum Linear Arrangement (MinLA) problem which disclose properties that can be exploited by Multi-Neighborhood Search (MNS) algorithms. As a foundation, we differentiate between swaps of labels attached to adjacent and non-adjacent nodes to create two new neighborhood classes, and show how our theorems yield efficient algorithms for updating key arrays used by local search procedures. In addition, we introduce a class of neighborhoods called set-based neighborhoods supported by a theorem that identifies solutions (labelings) for the MinLA problem in polynomial time that dominate exponential numbers of alternative solutions. The component neighborhoods within this new neighborhood class can be applied in various sequences in conjunction with the first two new neighborhoods introduced. Our results also apply to problems with objectives different than those of MinLA. Finally, our results make it possible to exploit the new neighborhoods according to the user's choice of MNS protocols and alternative local search algorithms.

The Bayesian Network based program dependence graph and its application to fault localization

Fault localization is an important and expensive task in software debugging. Some probabilistic graphical models such as probabilistic program dependence graph (PPDG) have been used in fault localization. However, PPDG is insufficient to reason across nonadjacent nodes and only support making inference about local anomaly. In this paper, we propose a novel probabilistic graphical model called Bayesian Network based Program Dependence Graph (BNPDG) that has the excellent inference capability for reasoning across nonadjacent nodes. We focus on applying the BNPDG to fault localization. Compared with the PPDG, our BNPDG-based fault localization approach overcomes the reasoning limitation across nonadjacent nodes and provides more precise fault localization by taking its output nodes as the common conditions to calculate the conditional probability of each non-output node. The experimental results show that our BNPDG-based fault localization approach can significantly improve the effectiveness of fault localization.

Distribution of FDG-avid nodes in esophageal cancer: implications for radiotherapy target delineation

PurposeClinical target volumes (CTV) for radiotherapy (RT) in esophageal cancer (EC) are based on standard expansions of primary tumor volume. Data is needed to define regions at highest risk for occult disease, based on histology and location of the primary tumor. We therefore reviewed PET scans in EC patients to characterize the location of FDG-avid lymph node metastases (LNM).Materials and methodsWe identified 473 EC patients with reviewable pre-treatment PET-CT scans. Tumors were classified by histology and location; 85% were distal or GE junction tumors and 71% were adenocarcinoma. FDG-avid LNM were classified using standard radiographic nodal atlases, and distances from primary tumor to paraesophageal LNM were also measured.ResultsThe most common LNM in upper EC were supraclavicular, retrotracheal and paratracheal. The most common LNM in lower EC were paraesophageal and in the gastrohepatic space. Overall, 55% of paraesophageal LNM were adjacent to primary tumor. Of upper esophageal tumors with paraesophageal LNM, 87% were adjacent to the tumor and none were >6 cm from tumor. However, 57% of lower esophageal tumors with paraesophageal LNM had non-adjacent paraesophageal nodes, 24% of which were >8 cm from the tumor.ConclusionA more data-driven and individualized approach to CTV delineation could improve the therapeutic ratio of RT in esophageal cancer. These results can guide CTV delineation by indicating the potential distribution of nodal involvement in esophageal cancer.Electronic supplementary materialThe online version of this article (doi:10.1186/s13014-016-0731-6) contains supplementary material, which is available to authorized users.

Identifying opinion leader nodes in online social networks with a new closeness evaluation algorithm

In online social networks, there are some influential opinion leader nodes who can be used to accelerate the spread of positive information and suppress the diffusion of rumors. If these opinion leaders can be identified timely and correctly, there will be contributing to guide the popular opinions. The closeness is introduced for mapping the relationship between the nodes according to the different interaction types in online social network. In order to measure the impact of the information transmission between non-adjacent nodes in online social networks, a closeness evaluating algorithm of the adjacent nodes and the non-adjacent nodes is given based on the relational features between users. By using the algorithm, the closeness between the adjacent nodes and the non-adjacent nodes can obtained depending on the interaction time of nodes and the delay of their hops. Furthermore, a more accurate and efficient betweenness centrality scheme based on the optimized algorithm with the degree of closeness and the corresponding updating strategy. The opinion leader nodes should be identified more accurately and efficiently under the improved algorithm because the considering of closeness between nodes in the network. Finally, the maximum spreading experiment is done for comparing the proposed method with other existing identifying opinion leader selecting schemes based on the Independent Cascade Model. The result of experiment shows the effectiveness and practicality of the evaluating algorithm.

Substrate-integrated waveguide filters based on mushroom-shaped resonators

This paper presents a new class of quasi-elliptic pass-band filters in substrate-integrated waveguide technology, which exhibits compact size and modular geometry. These filters are based on mushroom-shaped metallic resonators, and they can be easily implemented using a standard dual-layer printed circuit board manufacturing process. The presented filters exploit non-resonating modes to obtain coupling between non-adjacent nodes in the case of in-line geometry. The resulting structure is very compact and capable of transmission zeros. In this work, the singlet configuration is preliminarily investigated, and a parametric study is performed. The design of three-pole, four-pole, and higher-order filters is illustrated with examples and thoroughly discussed. A four-pole filter operating at the frequency of 4 GHz has been manufactured and experimentally verified, to validate the proposed technique.

Synthesis of Cross-Coupled Prototype Filters Including Resonant and Non-Resonant Nodes

In this paper, we present an approach to the synthesis of canonical cross-coupled prototype filters including resonant and non-resonant nodes (NRNs). The family of realizable topologies (including the folded and wheel configurations) is obtained through a recursive extraction technique and is characterized by all the couplings between adjacent nodes plus a number of cross-couplings between non-adjacent nodes (either resonant or non-resonant). The total number of distinct topologies that can be realized is ${\hbox{2}}^{({\rm np}-2)}$ with ${\rm np}$ order of the filter (i.e., number of poles). Some examples of synthesis are reported for illustrating the capabilities of the new proposed procedure. It is also shown how to de-normalize the synthesized low-pass prototype by means of the generalized coupling coefficients, which allow to extend the usual de-normalization procedure also to networks with NRNs.

Identifying node importance in complex networks

In this paper, we propose a novel node importance evaluation method from the perspective of the existence of mutual dependence among nodes. The node importance comprises its initial importance and the importance contributions from both the adjacent and non-adjacent nodes according to the dependence strength between them. From the simulation analyses on an example network and the ARPA network, we observe that our method can well identify the node importance. Then, the cascading failures on the Netscience and E-mail networks demonstrate that the networks are more vulnerable when continuously removing the important nodes identified by our method, which further proves the accuracy of our method.

Distribution of FDG-avid nodes in esophageal cancer: Implications for radiotherapy target delineation.

151 Background: Clinical target volumes (CTV) for radiation therapy (RT) in esophageal cancer (EC) are based on standard expansions of the primary tumor volume. Data is needed to better define the regions at highest risk for occult disease, based on histology and location of the primary tumor. We therefore reviewed PET scans in EC patients to characterize the frequency and location of FDG-avid lymph node metastases (LNM). Methods: We identified 474 EC patients with reviewable pre-treatment PET-CT scans who received RT. Tumors were classified by histology and location: upper esophageal (above carina, 15%), or lower/GE junction (85%). Locations of suspicious LNM were classified using standard radiographic nodal atlases, and distances from primary tumor to paraesophageal LNM were also measured. Results: FDG-avid LNM were identified and characterized in 204 patients. The most common LNM in upper EC were supraclavicular (1), retrotracheal (3P) and paratracheal (4). The most common LNM in lower EC were paraesophageal (8), and those in the gastrohepatic space (abdominal stations 1, 3 and 7). Overall, 55% of paraesophageal LNM were adjacent to the primary tumor. Of upper esophageal tumors with paraesophageal LNM, 87% were adjacent to the tumor and none were >6cm from tumor. However, 57% of lower esophageal tumors with paraesophageal LNM had non-adjacent paraesophageal nodes, half of which were >5cm from the tumor. Conclusions: A more data-driven and individualized approach to CTV delineation could improve the therapeutic ratio of RT in esophageal cancer. These results can guide CTV delineation by indicating the potential distribution of nodal involvement in esophageal cancer according to tumor location and histology. Because most paraesophageal LNM are adjacent to primary tumor, the necessity for uniformly large, longitudinal CTV expansions along the esophagus is unclear. [Table: see text]

Flaw-tolerance of nonlocal discrete systems and interpretation according to network theory

Discrete systems are modeled as a network of nodes (particles, molecules, or atoms) linked by nonlinear springs to simulate the action of van der Waals forces. Such systems are nonlocal if links connecting non-adjacent nodes are introduced. For their topological characterization, a nonlocality index (NLI) inspired by network theory is proposed. The mechanical response of 1D and 2D nonlocal discrete systems is predicted according to finite element (FE) simulations based on a nonlinear spring element for large displacements implemented in the FE programme FEAP. Uniaxial force-displacement responses of intact and defective systems (with links or nodes removed) are numerically simulated. Strain localization phenomena, size-scale effects and the ability to tolerate defects are investigated by varying the degree of nonlocality.

Simple Neural-Like P Systems for Maximal Independent Set Selection

Membrane systems (P systems) are distributed computing models inspired by living cells where a collection of processors jointly achieves a computing task. The problem of maximal independent set (MIS) selection in a graph is to choose a set of nonadjacent nodes to which no further nodes can be added. In this letter, we design a class of simple neural-like P systems to solve the MIS selection problem efficiently in a distributed way. This new class of systems possesses two features that are attractive for both distributed computing and membrane computing: first, the individual processors do not need any information about the overall size of the graph; second, they communicate using only one-bit messages.

Built-in Coloring for Highly-Concurrent Doubly-Linked Lists

This paper presents a novel approach for highly-concurrent nonblocking implementations of doubly-linked lists, based on dynamically maintaining a coloring of the nodes in the list. In these implementations, operations on non-adjacent nodes in the linked-list proceed without interfering with each other. Roughly speaking, the operations are implemented by acquiring nodes in the operation's data set, in the order of their colors, and then making the changes atomically. The length of waiting chains is restricted, thereby increasing concurrency, because the colors are taken from a small set. Operations carefully update the colors of the nodes they modify, so neighboring nodes in the list have different colors. A helping mechanism ensures progress in small neighborhoods of processes that keep taking steps. We use this approach in two new algorithms: CAS-Chromo uses an unary conditional primitive, cas, and allows insertions anywhere in the linked list and removals only at the ends, while DCAS-Chromo allows insertions and removals anywhere but uses a stronger primitive, dcas.

The Probability of Trojan Attacks on Multi-level Security Strategy based Network

The network information system based on multi-level security strategy (MLSS) is adopted by many organizations, as it reflects features of mandatory access control. Meanwhile, the network has also attracted a growing number of Trojan horses’ attacks. Considering the relationship between defend ability and security levels, the paper discusses that the Trojan horses’ attacks have a dependency on data path in the network and establishes three attacking probability models of single source node to single target node by single-path, single source node to single target node by multi-path and multi-source nodes to single target node by multi-path. Furthermore the paper describes the defend ability of the non-adjacent nodes with a new model. Finally the model is applied to a military network information system, and the results are reasonable.

Algorithms and formulations for the minimum cut separator problem

Abstract Given G = ( V , E ) an undirected graph and two specified nonadjacent nodes a and b of V, a cut separator is a subset F = δ ( C ) ⊆ E such that a , b ∈ V C and a and b belong to different connected components of the graph induced by V C . Given a nonnegative cost vector c ∈ R + | E | , cut separator of minimum cost. This new problem is closely related to the vertex separator problem. In this paper, we give a polynomial time algorithm for this problem. We also present four equivalent linear formulations, and we show their tightness. Using these results we obtain an explicit short polyhedral description of the dominant of the cut separator polytope.

Optimal constrained graph exploration

We address the problem of constrained exploration of an unknown graph G = ( V , E ) from a given start node s with either a tethered robot or a robot with a fuel tank of limited capacity, the former being a tighter constraint. In both variations of the problem, the robot can only move along the edges of the graph, for example, it cannot jump between nonadjacent nodes. In the tethered robot case, if the tether (rope) has length l , then the robot must remain within distance l from the start node s . In the second variation, a fuel tank of limited capacity forces the robot to return to s after traversing C edges. The efficiency of algorithms for both variations of the problem is measured by the number of edges traversed during the exploration. We present an algorithm for a tethered robot that explores the graph in Θ(| E |) edge traversals. The problem of exploration using a robot with a limited fuel tank capacity can be solved with a simple reduction from the tethered robot case and also yields a Θ(| E |) algorithm. This improves on the previous best-known bound of O (| E | + | V |log 2 | V |). Since the lower bound for the graph exploration problems is Ω(| E |), our algorithm is optimal within a constant factor.