Large-scale Complex Networks Research Articles

Parametric Optimal Design System of Construction System Based on Distributed Optimal Algorithm

With the rapid development of science and technology and the continuous expansion of network scale, traditional centralized control and optimization techniques have been difficult to solve large-scale complex network problems, and distributed optimization algorithms with more robustness and flexibility have attracted more and more attention. In view of the irreplaceable advantages of multi-intelligent systems in distributed computing, many researchers have used them as carriers of distributed optimization for theoretical research and application promotion. For the distributed optimization problem in the undirected network environment, this article outlines and summarizes the existing gradient-based distributed simultaneous optimization algorithms, based on which a novel distributed simultaneous momentum acceleration optimization algorithm is proposed. Under the premise that the local objective function is strongly convex and Lipschitz continuous, the algorithm uses heterogeneous steps to precisely drive each node to converge asymptotically to the global optimal solution, and we reduce or compensate the effect of the original error on the pairwise error by introducing an adjustable symmetric matrix. The distributed resource allocation problem under undirected graphs is studied, in which the local cost function of the intelligences is unknown. Using sinusoidal signal excitation, the input and output of the cost function are utilized, and the first-order and second-order polar search algorithms are designed, respectively. In addition, the algorithm innovatively introduces the dual acceleration mechanism combining the Nesterov gradient method and the Heavy-Ball method to greatly improve the convergence speed of the algorithm. Based on the interrelationship of the agreement error, the optimal distance, the state difference, and the tracking error, the step size and the range of the momentum parameters for the linear convergence of the algorithm to the optimal solution are analyzed using the linear matrix inequality technique.

A high-speed railway network dataset from train operation records and weather data

High-speed train operation data are reliable and rich resources in data-driven research. However, the data released by railway companies are poorly organized and not comprehensive enough to be applied directly and effectively. A public high-speed railway network dataset suitable for research is still lacking. To support the research in large-scale complex network, complex dynamic system and intelligent transportation, we develop a high-speed railway network dataset, containing the train operation data in different directions from October 8, 2019 to January 27, 2020, the train delay data of the railway stations, the junction stations data, and the mileage data of adjacent stations. In the dataset, weather, temperature, wind power and major holidays are considered as factors affecting train operation. Potential research values of the dataset include but are not limited to complex dynamic system pattern mining, community detection and discovery, and train delay analysis. Besides, the dataset can be used to solve various railway operation and management problems, such as passenger service network improvement, train real-time dispatching and intelligent driving assistance.

Searching for Best Network Topologies with Optimal Synchronizability: A Brief Review

The Laplacian eigenvalue spectrum of a complex network contains a great deal of information about the network topology and dynamics, particularly affecting the network synchronization process and performance. This article briefly reviews the recent progress in the studies of network synchronizability, regarding its spectral criteria and topological optimization, and explores the role of higher-order topologies in measuring the optimal synchronizability of large-scale complex networks.

An improved ant-based heuristic approach for solving the longest cycle problem in large-scale complex networks

An improved ant-based heuristic approach for solving the longest cycle problem in large-scale complex networks

Accessibility-oriented performance evaluation of high-speed railways using a three-layer network model

High-speed railways, as dominant transportation patterns, constitute a large-scale complex network. Network performance is critical for high-level transportation services in transportation management. However, existing studies generally adopt pure topological indicators to evaluate performance in a single network and without considering the specific characteristics of rail transportation. This paper modeled the high-speed railway as a three-layer network including topological, functional, and service layers and assessed the integrated network performance from the view of transportation accessibility. In addition, the rail stations were prioritized and their performances in different layers were compared to analyze whether the total integrated performance was restricted by its connection abilities, functional capacities, and transportation service level. Finally, a case study of the high-speed rail network in China and comparisons with other traditional evaluation methods were conducted to verify the feasibility and superiority of the model. The findings of this study enable to provide management implications for rail transportation planning, organization, and operations.

Identifying Influential Nodes in Complex Networks Based on Multiple Local Attributes and Information Entropy.

Identifying influential nodes in complex networks has attracted the attention of many researchers in recent years. However, due to the high time complexity, methods based on global attributes have become unsuitable for large-scale complex networks. In addition, compared with methods considering only a single attribute, considering multiple attributes can enhance the performance of the method used. Therefore, this paper proposes a new multiple local attributes-weighted centrality (LWC) based on information entropy, combining degree and clustering coefficient; both one-step and two-step neighborhood information are considered for evaluating the influence of nodes and identifying influential nodes in complex networks. Firstly, the influence of a node in a complex network is divided into direct influence and indirect influence. The degree and clustering coefficient are selected as direct influence measures. Secondly, based on the two direct influence measures, we define two indirect influence measures: two-hop degree and two-hop clustering coefficient. Then, the information entropy is used to weight the above four influence measures, and the LWC of each node is obtained by calculating the weighted sum of these measures. Finally, all the nodes are ranked based on the value of the LWC, and the influential nodes can be identified. The proposed LWC method is applied to identify influential nodes in four real-world networks and is compared with five well-known methods. The experimental results demonstrate the good performance of the proposed method on discrimination capability and accuracy.

Adaptive parallel and distributed simulation of complex networks

Complex networks are an important methodology to model several (if not all) aspects of the real world, in which multiple entities interact, in some way. While many aspects related to such interactions can be investigated by looking at the general mathematical metrics of the networks, an alternative approach lies in the simulation of some application protocol on top of (large scale) complex networks. In this paper, we present a study on this intricate problem. The complexity of the simulation is due to the need to model all the interactions among network nodes. We focus on discrete-event simulation, a simulation methodology that enables both sequential (i.e. monolithic) and Parallel And Distributed Simulation (i.e. PADS) approaches. We discuss the performance and scalability requirements that the simulator should have. We also introduce a case study based on the agent-based simulation of gossip dissemination on top of a complex network. To demonstrate the viability of this simulation technique, we focus on a tool we built to simulate complex networks. The tool exploits adaptive partitioning mechanisms, which are essential to reduce the communication overhead in the PADS. An experimental evaluation has been conducted using different network topologies and simulator setups. Results demonstrate the feasibility of the approach to simulate complex networks.

A novel metamodel-based framework for large-scale dynamic origin–destination demand calibration

Calibrating dynamic traffic demand for stochastic traffic simulators is one of the big challenges due to computational burden. This paper proposes a novel framework to calibrate a dynamic car origin–destination matrix of large-scale networks, which has high computational efficiency. The proposed framework is relied on the metamodel optimization technique, based on the reference of aggregated traffic flow dynamics as expressed by the recent advance in traffic flow theory, namely the bi-modal macroscopic fundamental diagram (MFD).We validate our proposed approach with the Stochastic Perturbation Simultaneous Approximation (SPSA) algorithm which is widely used for the same purpose in the literature. Our result confirms that our approach can facilitate the demand calibration effectively. We also show that the traffic conditions at a link level are also reproduced realistically from our network-level calibration, and that our approach outperforms the traditional link-level OD calibration by comparing with the Aimsun OD adjustment in addition to SPSA.Furthermore, we demonstrate via different realistic network studies that the proposed approach is computationally efficient compared to the existing state-of-the-art approach. Our result shows that the proposed approach can be applied effectively regardless of the topology of the network, and that the model parameters do not have significant influence on the optimization results. Our calibration results show that just a few iterations are needed to calibrate the OD demand in the proposed approach even for the large-scale complex network underpinned by a bi-modal MFD derived from multiple real data sources.

Deep Link Entropy for Quantifying Edge Significance in Social Networks

Through online political communications, fragmented groups appear around ideological lines, which might form echo chambers if the communications within like-minded groups are dominant over the communications among different-minded groups, potentially contributing to political polarization and extremism. The antidote is the interactions between individuals who constitute social bridges between different minded groups. Hence, exploring the significance of connections between the individuals of a network is a center of attraction especially for the global connectivity and diffusion in networks. Based on the divergence of probability distributions of pairs of nodes, Link Entropy (LE) is a recently proposed method outperforming the others in quantifying edge significance. In this work, considering that the adjacent nodes of the two nodes of an edge are also in charge in determining its significance, we propose the Deep Link Entropy (DLE) method for a more precise quantification through taking into account the uncertainty distributions of the adjacent nodes as well. We show experimentally that DLE significantly outperforms LE especially in large-scale complex network with several groups or communities. We believe our method contributes to not only online political communications but a wide range of fields from biology to quantum networks, where edge significance has an operational meaning.

Structure inference of networked system with the synergy of deep residual network and fully connected layer network

The networked systems are booming in multi-disciplines, including the industrial engineering system, the social system, and so on. The network structure is a prerequisite for the understanding and exploration of networked systems. However, the network structure is always unknown in practice, thus, it is significant yet challenging to investigate the inference of network structure. Although some model-based methods and data-driven methods, such as the phase-space based method and the compressive sensing based method, have investigated the structure inference tasks, they were time-consuming due to the greedy iterative optimization procedure, which makes them difficult to satisfy real-time structure inference requirements. Although the reconstruction time of L1 and other methods is short, the reconstruction accuracy is very low. Inspired by the powerful representation ability and time efficiency for the structure inference with the deep learning framework, a novel synergy method combines the deep residual network and fully connected layer network to solve the network structure inference task efficiently and accurately. This method perfectly solves the problems of long reconstruction time and low accuracy of traditional methods. Moreover, the proposed method can also fulfill the inference task of large scale complex network, which further indicates the scalability of the proposed method.

Synthesizing Brain-network-inspired Interconnections for Large-scale Network-on-chips

Brain network is a large-scale complex network with scale-free, small-world, and modularity properties, which largely supports this high-efficiency massive system. In this article, we propose to synthesize brain-network-inspired interconnections for large-scale network-on-chips. First, we propose a method to generate brain-network-inspired topologies with limited scale-free and power-law small-world properties, which have a low total link length and extremely low average hop count approximately proportional to the logarithm of the network size. In addition, given the large-scale applications, considering the modularity of the brain-network-inspired topologies, we present an application mapping method, including task mapping and deterministic deadlock-free routing, to minimize the power consumption and hop count. Finally, a cycle-accurate simulator BookSim2 is used to validate the architecture performance with different synthetic traffic patterns and large-scale test cases, including real-world communication networks for the graph processing application. Experiments show that, compared with other topologies and methods, the brain-network-inspired network-on-chips (NoCs) generated by the proposed method present significantly lower average hop count and lower average latency. Especially in graph processing applications with a power-law and tightly coupled inter-core communication, the brain-network-inspired NoC has up to 70% lower average hop count and 75% lower average latency than mesh-based NoCs.

Big Data Processing on Volunteer Computing

In order to calculate the node big data contained in complex networks and realize the efficient calculation of complex networks, based on voluntary computing, taking ICE middleware as the communication medium, the loose coupling distributed framework DCBV based on voluntary computing is proposed. Then, the Master, Worker, and MiddleWare layers in the framework, and the development structure of a DCBV framework are designed. The task allocation and recovery strategy, message passing and communication mode, and fault tolerance processing are discussed. Finally, to calculate and verify parameters such as the average shortest path of the framework and shorten calculation time, an improved accurate shortest path algorithm, the N-SPFA algorithm, is proposed. Under different datasets, the node calculation and performance of the N-SPFA algorithm are explored. The algorithm is compared with four approximate shortest-path algorithms: Combined Link and Attribute (CLA), Lexicographic Breadth First Search (LBFS), Approximate algorithm of shortest path length based on center distance of area division (CDZ), and Hub Vertex of area and Core Expressway (HEA-CE). The results show that when the number of CPU threads is 4, the computation time of the DCBV framework is the shortest (514.63 ms). As the number of CPU cores increases, the overall computation time of the framework decreases gradually. For every 2 additional CPU cores, the number of tasks increases by 1. When the number of Worker nodes is 8 and the number of nodes is 1, the computation time of the framework is the shortest (210,979 ms), and the IO statistics data increase with the increase of Worker nodes. When the datasets are Undirected01 and Undirected02, the computation time of the N-SPFA algorithm is the shortest, which is 4520 ms and 7324 ms, respectively. However, the calculation time in the ca-condmat_undirected dataset is 175,292 ms, and the performance is slightly worse. Overall, however, the performance of the N-SPFA and SPFA algorithms is good. Therefore, the two algorithms are combined. For networks with less complexity, the computational scale coefficient of the SPFA algorithm can be set to 0.06, and for general networks, 0.2. When compared with other algorithms in different datasets, the pretreatment time, average query time, and overall query time of N-SPFA algorithm are the shortest, being 49.67 ms, 5.12 ms, and 94,720 ms, respectively. The accuracy (1.0087) and error rate (0.024) are also the best. In conclusion, voluntary computing can be applied to the processing of big data, which has a good reference significance for the distributed analysis of large-scale complex networks.

A parallel multi-objective evolutionary algorithm for community detection in large-scale complex networks

Community detection in large-scale complex networks has recently received significant attention as the volume of available data is becoming larger. The use of evolutionary algorithms (EAs) for community detection in large-scale networks has gained considerable popularity because these algorithms are fairly effective in networks with a relatively small number of nodes. In this paper, we propose a parallel multi-objective EA, called PMOEA, for community detection in large-scale networks, where the communities associated with key network nodes are detected in parallel. Specifically, we develop a multi-objective and a single-objective EA. The former is used to detect the communities of a key node instead of all communities in the network. The latter obtains the communities in the entire network using the previously detected communities of each key node. The performance of the proposed method was verified on both large-scale synthetic benchmark networks and real-world networks. The results demonstrated the superiority of PMOEA over six EA-based and two non-EA-based community-detection algorithms for large-scale networks.

PLDLS: A novel parallel label diffusion and label Selection-based community detection algorithm based on Spark in social networks

Parallel and distributed community detection in large-scale complex networks, such as social networks, is a challenging task. Parallel and distributed algorithm with high accuracy and low computational complexity is one of the essential issues in the community detection field. In this paper, we propose a novel fast, and accurate Spark-based parallel label diffusion and label selection-based (PLDLS) community detection algorithm with two-step of label diffusion of core nodes along with a new label selection (propagation) method. We have used multi-factor criteria for computing node's importance and adopted a new method for selecting core nodes. In the first phase, utilizing the fact that nodes forming triangles, tend to be in the same community, parallel label diffusion of core nodes is performed to diffuse labels up to two levels. In the second phase, through an iterative and parallel process, the most appropriate labels are assigned to the remaining nodes. PLDLS proposes an improved robust version of LPA by putting aside randomness parameter tuning. Furthermore, we utilize a fast and parallel merge phase to get even more dense and accurate communities. Conducted experiments on real-world and artificial networks, indicates the better accuracy and low execution time of PLDLS in comparison with other examined methods.

Modified Lomax model: a heavy-tailed distribution for fitting large-scale real-world complex networks

Real-world networks are generally claimed to be scale-free, meaning that the degree distributions follow the classical power-law, at least asymptotically. Yet, closer observation shows that the classical power-law distribution is often inadequate to meet the data characteristics due to the existence of a clearly identifiable non-linearity in the entire degree distribution in the log-log scale. The present paper proposes a new variant of the popular heavy-tailed Lomax distribution which we named as the Modified Lomax (MLM) distribution that can efficiently capture the crucial aspect of heavy-tailed behavior of the entire degree distribution of real-world complex networks. The proposed MLM model, derived from a hierarchical family of Lomax distributions, can efficiently fit the entire degree distribution of real-world networks without removing lower degree nodes as opposed to the classical power-law based fitting. The MLM distribution belongs to the maximum domain of attraction of the Frechet distribution and is right tail equivalent to Pareto distribution. Various statistical properties including characteristics of the maximum likelihood estimates and asymptotic distributions have also been derived for the proposed MLM model. Finally, the effectiveness of the proposed MLM model is demonstrated through rigorous experiments over fifty real-world complex networks from diverse applied domains.

Distributed adaptive stabilization

In this paper we consider distributed adaptive stabilization for uncertain multivariable linear systems with a time-varying diagonal matrix gain. We show that uncertain multivariable linear systems are stabilizable by diagonal matrix high gains if the system matrix is an H-matrix with positive diagonal entries. Based on matrix measure and stability theory for diagonally dominant systems, we consider two classes of uncertain linear systems, and derive a threshold condition to ensure their exponential stability by a monotonically increasing diagonal gain matrix. When each individual gain function in the matrix gain is updated by state-dependent functions using only local state information, the boundedness and convergence of both system states and adaptive matrix gains are guaranteed. We apply the adaptive distributed stabilization approach to adaptive synchronization control for large-scale complex networks consisting of nonlinear node dynamics and time-varying coupling weights. A unified framework for adaptive synchronization is proposed that includes several general design approaches for adaptive coupling weights to guarantee network synchronization.

Computationally efficient sandbox algorithm for multifractal analysis of large-scale complex networks with tens of millions of nodes.

Among various algorithms of multifractal analysis (MFA) for complex networks, the sandbox MFA algorithm behaves with the best computational efficiency. However, the existing sandbox algorithm is still computationally expensive for MFA of large-scale networks with tens of millions of nodes. It is also not clear whether MFA results can be improved by a largely increased size of a theoretical network. To tackle these challenges, a computationally efficient sandbox algorithm (CESA) is presented in this paper for MFA of large-scale networks. Distinct from the existing sandbox algorithm that uses the shortest-path distance matrix to obtain the required information for MFA of networks, our CESA employs the compressed sparse row format of the adjacency matrix and the breadth-first search technique to directly search the neighbor nodes of each layer of center nodes, and then to retrieve the required information. A theoretical analysis reveals that the CESA reduces the time complexity of the existing sandbox algorithm from cubic to quadratic, and also improves the space complexity from quadratic to linear. Then the CESA is demonstrated to be effective, efficient, and feasible through the MFA results of (u,v)-flower model networks from the fifth to the 12th generations. It enables us to study the multifractality of networks of the size of about 11 million nodes with a normal desktop computer. Furthermore, we have also found that increasing the size of (u,v)-flower model network does improve the accuracy of MFA results. Finally, our CESA is applied to a few typical real-world networks of large scale.

Network Risk Propagation Model Based On Boundary-Center-Point-Search Algorithm

Because the power CPS network with community structure has the unique structural characteristics of internal nodes and boundary nodes, the boundary nodes are more likely to cause cascading failures when attacked, which plays an extremely important role in network risk fault propagation. In this paper, the Boundary node is regarded as an important immune object, and these key node targets are quickly identified to take immune measures. A Boundary center-point-search (BCS) algorithm is proposed. According to the disadvantage of the original risk propagation model which ignores the fact that normal nodes can get immunity through artificial means, a risk propagation model is proposed. Compared with other local immune algorithms, the proposed BCS immune algorithm not only can accurately identify the boundary center points between communities, but also has a faster computational speed, which can provide reference for risk propagation control and prevention in large-scale complex networks with incomplete structural information.

The detection of low-rate DoS attacks using the SADBSCAN algorithm

Low-rate denial-of-service (DoS) attacks, which can exploit vulnerabilities in Internet protocols to deteriorate the quality of service, are variants of DoS attacks. It is challenging to identify low-rate DoS attacks using traditional DoS defence mechanisms due to their low attack rate and stealthy nature. Most of the existing attack detection techniques are based on statistical analysis and signal processing. They usually show a high false negative rate and are only applicable to small-scale data. We propose a new low-rate DoS attack detection scheme based on the self-adaptive density-based spatial clustering of applications with noise (SADBSCAN) algorithm. The SADBSCAN algorithm provides a solution to adaptively identify clusters in multidensity datasets. We use the SADBSCAN algorithm to group network traffic according to the characteristics of the network traffic subject to low-rate DoS attacks. Then, we use cosine similarity to determine whether the groups contain low-rate DoS attacks. To evaluate performance, we conducted experiments and compared the results with those of other detection solutions. The experimental data include data generated by the NS-2 and TestBed simulations and the WIDE public dataset. The results show that our scheme improves the detection accuracy, reduces the false negative rate, and can be adapted to large-scale complex network environments.

Iterative multi‐scale residual network for deblurring

In dynamic scene deblurring, recent neural network–based methods have been very successful. But with the improvement of deep deblurring performance, network structure and learning become more complicated. Compared with large-scale network parameters and complex network structures, an iterative multi-scale residual network to achieve a more effective parameter sharing scheme is proposed. In each iterative unit, fast multi-scale residual blocks to replace superimposed convolutional layers or classic residual blocks are used. On the basis of preventing model overfitting, the receptive field of the network is increased. At the same time, the gated recurrent unit is introduced to connect modules of different stages. The model does not rely on the estimation of the blur kernel and directly generates sharp images in an end-to-end manner. The experimental structure on the benchmark dataset and real-world images showed that this method has better quality than the existing methods in terms of large-scale blur and subjective perception effects, both in quantitative and qualitative terms.