Real Network Data Research Articles

ViCTS: A novel network partition algorithm for scalable agent-based modeling of mass evacuation

Emergency evacuation is a critical response to deadly disasters such as hurricanes, floods, and earthquakes, etc. However, mass emergency evacuation itself is a complex process that sometimes could lead to chaotic situations and unintended consequences. In many emergency scenarios, mass evacuation is necessary to cope with severe public threats within tight spatiotemporal ranges. To better understand complex phenomena like mass evacuation, and study possible consequences, agent-based models (ABMs) have been widely developed in previous work. Existing models simulate individual behaviors, posing computational challenges when applied to large geographic areas and sophisticated behaviors. A key strategy for resolving such computational challenges is to partition transportation networks into smaller regions and resolve corresponding computational costs by taking advantage of advanced cyberinfrastructure and cyberGIS. In this study, a novel network partition algorithm is developed to improve the scalability of agent-based modeling of mass evacuation based on a cutting-edge cyberGIS-enabled computational framework that exploits the spatial movement patterns of emergency evacuation. Specifically, the algorithm is termed as Voronoi Clustering based on Target-Shift, or ViCTS. It is enlightened by network Voronoi diagrams and designed to resolve computational scalability challenges caused by the unique characteristics of evacuation traffic. We conducted a set of computational experiments with real street network data in various evacuation scenarios to test the effectiveness and efficiency of the algorithm. Computational experiments show that ViCTS outperforms a widely used network partition algorithm for microscopic traffic simulation in terms of achieving optimal computational performance by balancing computational loads and reducing communications across high-performance parallel computing resources.

Overlapping community detection for count-value networks

Detecting network overlapping community has become a very hot research topic in the literature. However, overlapping community detection for count-value networks that naturally arise and are pervasive in our modern life, has not yet been thoroughly studied. We propose a generative model for count-value networks with overlapping community structure and use the Indian buffet process to model the community assignment matrix Z; thus, provide a flexible nonparametric Bayesian scheme that can allow the number of communities K to increase as more and more data are encountered instead of to be fixed in advance. Both collapsed and uncollapsed Gibbs sampler for the generative model have been derived. We conduct extensive experiments on simulated network data and real network data, and estimate the inference quality on single variable parameters. We find that the proposed model and inference procedure can bring us the desired experimental results.

Linear Power Flow Method Improved With Numerical Analysis Techniques Applied to a Very Large Network

In this paper, we propose a fast linear power flow method using a constant impedance load model to simulate both the entire Low Voltage (LV) and Medium Voltage (MV) networks in a single simulation. Accuracy and efficiency of this linear approach are validated by comparing it with the Newton power flow algorithm and a commercial network design tool Vision on various distribution networks including real network data. Results show that our method can be as accurate as classical Nonlinear Power Flow (NPF) methods using a constant power load model and additionally, it is much faster than NPF computations. In our research, it is shown that voltage problems can be identified more efficiently when MV and LV are integrally evaluated. Moreover, Numerical Analysis (NA) techniques are applied to the Large Linear Power Flow (LLPF) problem with 27 million nonzeros in order to improve the computation time by studying the properties of the linear system. Finally, the original computation times of LLPF problems with real and complex components are reduced by 2.8 times and 5.7 times, respectively.

Classes of preferential attachment and triangle preferential attachment models with power-law spectra

AbstractPreferential attachment (PA) models are a common class of graph models which have been used to explain why power-law distributions appear in the degree sequences of real network data. Among other properties of real-world networks, they commonly have non-trivial clustering coefficients due to an abundance of triangles as well as power laws in the eigenvalue spectra. Although there are triangle PA models and eigenvalue power laws in specific PA constructions, there are no results that existing constructions have both. In this article, we present a specific Triangle Generalized Preferential Attachment Model that, by construction, has non-trivial clustering. We further prove that this model has a power law in both the degree distribution and eigenvalue spectra.

Asymptotic distribution of modularity in networks

The structure of complex networks is an important aspect in the study of the real network data. Quite often, it is desirable to know the division of the network into communities. A large number of community detection algorithms have been proposed to probe the community structure of complex networks. For a specific partition of a given network, we show that the distribution of modularity under a null hypothesis of free labeling is asymptotically normal when the size of the network gets large. The significance of the partition is defined based on this asymptotic distribution, which can help assess its goodness. Two different partitions can also be compared statistically. Simulation studies and real data analyses are performed for illustration.

INSTRUMENTATION IN SEISMOLOGICAL OBSERVATORY – SKOPJE MONITORING AND DATA PROCESSING

The present activities in the field of earthquake seismology in Republic of North Macedonia are carried out by the Seismological Observatory at the Faculty of Natural Sciences and Mathematics, University "Sts. Cyril and Methodius" - Skopje, founded in 1957. Seismological Observatory in Skopje with telemetric network of digital seismological stations systematically monitors the seismic activity in the territory of Republic of North Macedonia and the bordering areas and also records the regional and teleseismic earthquakes. The instrumental seismological data at the Seismological Observatory in Skopje and at stations of its network have been always obtained by instrumentation which had followed the world trends. Actual scientific methods and, lately, the most sophisticated computer softwares have been used in analyses. Earthquakes with local magnitudes down to zero can be recorded and analyzed with the latest instrumentation by real time telemetric network data exchange and used softwares. The predominant hypocentral depths are less than 15 km. These data give opportunity for epicentral areas predefinition of the on the territory of Republic of North Macedonia, calculation of many source parameters and the mean velocity of the relative tectonic movement of the fault blocks, as well.

Convex Optimization for Group Feature Selection in Networked Data

Feature selection is at the heart of machine learning, and it is effective at facilitating data interpretability and improving prediction performance by defying the curse of dimensionality. Group feature selection is often used to reveal relationships in structured data and provide better predictive power compared with the standard feature selection methods without consideration of the grouped structure. We study a group feature selection problem in networked data in which edge weights are considered as features, while each node in the network is regarded as a group feature. This problem is particularly useful in feature selection for neuroimaging data, where the data are high dimensional and the intrinsic networked structure among the features (i.e., connectivities between regions) in brain data has to be captured properly. We propose a mathematical model based on the support vector machines (SVM), which entails the ℓ0norm regularization to restrict the number of nodes (i.e., groups). To cope with the computational challenge of the ℓ0norm regularization, we develop a convex relaxation reformulation of the proposed model as a convex semiinfinite programming (SIP). We then introduce a new iterative algorithm that achieves an optimal solution for this convex SIP. Experimental results for synthetic and real brain network data sets show that our approach gives better predictive performance compared with the state-of-the-art group feature selection and the standard feature selection methods. Our technique additionally yields a sparse subnetwork solution that is easier to interpret than those obtained by other methods.

TransConv: Relationship Embedding in Social Networks

Representation learning (RL) for social networks facilitates real-world tasks such as visualization, link prediction and friend recommendation. Traditional knowledge graph embedding models learn continuous low-dimensional embedding of entities and relations. However, when applied to social networks, existing approaches do not consider the rich textual communications between users, which contains valuable information to describe social relationships. In this paper, we propose TransConv, a novel approach that incorporates textual interactions between pair of users to improve representation learning of both users and relationships. Our experiments on real social network data show TransConv learns better user and relationship embeddings compared to other state-of-theart knowledge graph embedding models. Moreover, the results illustrate that our model is more robust for sparse relationships where there are fewer examples.

Node Attributed Query Access Algorithm Based on Improved Personalized Differential Privacy Protection in Social Network

The existing differential privacy for social network graphs data published method is mainly focused on graph synthesis. But the privacy budget is set by data owner without adequate consideration of the differences in privacy requirements between individual users. And it published a topology data of social network data, which does not combine the independent attribute information and attribute information of the individual user with the correlation of edge information. This paper researches the influence of both social network attribute graph node properties and the correlation of edge information under the condition of considering the user privacy requirements, thus a social network attributes graphs algorithm is proposed under personalized differential privacy, which is for the independent attribute information between users. Since the node properties do not match the data type, our proposed mode will make the data partitioning in the data set, and it also divides and calculates the probability according to the node attribute distribution query function. Finally, through two real social network data sets, our proposed algorithms will execute for experimental comparison, and are verified they validity and usability through the experimental results.

Clustering‐based real‐time anomaly detection—A breakthrough in big data technologies

AbstractOff late, the ever increasing usage of a connected Internet‐of‐Things devices has consequently augmented the volume of real‐time network data with high velocity. At the same time, threats on networks become inevitable; hence, identifying anomalies in real time network data has become crucial. To date, most of the existing anomaly detection approaches focus mainly on machine learning techniques for batch processing. Meanwhile, detection approaches which focus on the real‐time analytics somehow deficient in its detection accuracy while consuming higher memory and longer execution time. As such, this paper proposes a novel framework which focuses on real‐time anomaly detection based on big data technologies. In addition, this paper has also developed streaming sliding window local outlier factor coreset clustering algorithms (SSWLOFCC), which was then implemented into the framework. The proposed framework that comprises BroIDS, Flume, Kafka, Spark streaming, SparkMLlib, Matplot and HBase was evaluated to substantiate its efficacy, particularly in terms of accuracy, memory consumption, and execution time. The evaluation is done by performing critical comparative analysis using existing approaches, such as K‐means, hierarchical density‐based spatial clustering of applications with noise (HDBSCAN), isolation forest, spectral clustering and agglomerative clustering. Moreover, Adjusted Rand Index and memory profiler package were used for the evaluation of the proposed framework against the existing approaches. The outcome of the evaluation has substantially proven the efficacy of the proposed framework with a much higher accuracy rate of 96.51% when compared to other algorithms. Besides, the proposed framework also outperformed the existing algorithms in terms of lesser memory consumption and execution time. Ultimately the proposed solution enable analysts to precisely track and detect anomalies in real time.

Multilayer overlapping community detection using multi-objective optimization

Abstract Community detection in multilayer networks such as social or information networks, due to its vast practical applications has attracted many attentions in the past years. Although some researches have been done to develop monoplex methods to multilayers, but because of the complexity of multilayer networks they are in their infancy. In this study, initially, the definition of community in single layer networks is extended and a new definition for multilayer community is presented. Then, regarding the importance of overlapping communities in real networks, a comprehensive definition for overlapping multilayer community is presented. Furthermore, in order to address the problem of multilayer community detection, a two-phase approach has been adopted. In the first phase, a multi-objective mathematical model is developed, and optimized using a genetic algorithm based method (NSGA-II) to achieve a set of solutions (Pareto front). Each potential solution represents a partition of the multilayer network. In the second phase, the best solution is chosen from the Pareto fronts, using a novel algorithm. This proposed algorithm, handles the obstacles of genetic representation and is able to detect the optimal multilayer overlapping communities. Experiments on both synthetic mLFR networks with different parameters and six real data networks show the performance of the algorithm in terms of NMI and multilayer modularity.

A novel complex network community detection approach using discrete particle swarm optimization with particle diversity and mutation

Community detection in complex network is a hot issue in the field of complex network. This paper proposed a community detection approach in complex network using discrete particle swarm optimization with particle diversity and mutation (DPSO-PDM) strategy, aiming at the modularity optimization. Firstly, this paper redefined particle coding, particle velocity, particle position and evolutionary operation in discretization, which effectively solve the problem of traditional evolutionary algorithm in need of prior knowledge. Secondly, considering the traditional particle swarm optimization algorithm is easy to be trapped in the local optimum; this paper used the hybridizing inertia weight adjustment strategy based on new particle diversity and adaptive mutation strategy to avoid the local convergence of the algorithm. Finally this paper applied the DPSO-PDM algorithm to artificial baseline network data sets and real network data sets. Theoretic analysis and experimental results have shown that DPSO-PDM is effective to detect community structure with stable community division quality and global convergence in complex network.

An Advanced Algorithm for Higher Network Navigation in Social Internet of Things Using Small-World Networks

The Internet of Things (IoT) is a recent evolutionary technology that has been the primary focus of researchers for the last two decades. In the IoT, an enormous number of objects are connected together using diverse communications protocols. As a result of this massive object connectivity, a search for the exact service from an object is difficult, and hence the issue of scalability arises. In order to resolve this issue, the idea of integrating the social networking concept into the IoT, generally referred as the Social Internet of Things (SIoT) was introduced. The SIoT is gaining popularity and attracting the attention of the research community due to its flexible and spacious nature. In the SIoT, objects have the ability to find a desired service in a distributed manner by using their neighbors. Although the SIoT technique has been proven to be efficient, heterogeneous devices are growing so exponentially that problems can exist in the search for the right object or service from a huge number of devices. In order to better analyze the performance of services in an SIoT domain, there is a need to impose a certain set of rules on these objects. Our novel contribution in this study is to address the link selection problem in the SIoT by proposing an algorithm that follows the key properties of navigability in small-world networks, such as clustering coefficients, path lengths, and giant components. Our algorithm empowers object navigability in the SIoT by restricting the number of connections for objects, eliminating old links or having fewer connections. We performed an extensive series of experiments by using real network data sets from social networking sites like Brightkite and Facebook. The expected results demonstrate that our algorithm is efficient, especially in terms of reducing path length and increasing the average clustering coefficient. Finally, it reflects overall results in terms of achieving easier network navigation. Our algorithm can easily be applied to a single node or even an entire network.

On community detection in complex networks based on different training algorithms: A case study on prediction of depression of internet addiction

Community structure is an important feature of complex networks. In recent years, community detection algorithms based on optimization has been of interest for many researchers. One way to detect these communities is the use of algorithms based on swarm intelligence to find the optimal solution. Cuckoo optimization is discussed, and a new objective function is presented. The proposed method tries to maximize network modularity function and the similarity of nodes to each other at the same time. It also seeks to provide a better equation to calculate the similarity of nodes in a complex network. New objective function has raised the speed of convergence to the optimal solution and provides a solution with better quality. The results of simulations conducted on a real network data set show that the proposed method discovers communities with acceptable and efficient quality. The proposed methods are tested for prediction of depression of internet addiction and corresponding results are observed.

Multi-Objective Optimization-Based Networked Multi-Label Active Learning

Along with the fast development of network applications, network research has attracted more and more attention, where one of the most important research directions is networked multi-label classification. Based on it, unknown labels of nodes can be inferred by known labels of nodes in the neighborhood. As both the scale and complexity of networks are increasing, the problems of previously neglected system overhead are turning more and more seriously. In this article, a novel multi-objective optimization-based networked multi-label seed node selection algorithm (named as MOSS) is proposed to improve both the prediction accuracy for unknown labels of nodes from labels of seed nodes during classification and the system overhead for mining the labels of seed nodes with third parties before classification. Compared with other algorithms on several real networked data sets, MOSS algorithm not only greatly reduces the system overhead before classification but also improves the prediction accuracy during classification.

Percolation and the Effective Structure of Complex Networks

Analytical approaches to model the structure of complex networks can be distinguished into two groups according to whether they consider an intensive (e.g., fixed degree sequence and random otherwise) or an extensive (e.g., adjacency matrix) description of the network structure. While extensive approaches---such as the state-of-the-art Message Passing Approach---typically yield more accurate predictions, intensive approaches provide crucial insights on the role played by any given structural property in the outcome of dynamical processes. Here we introduce an intensive description that yields almost identical predictions to the ones obtained with MPA for bond percolation. Our approach distinguishes nodes according to two simple statistics: their degree and their position in the core-periphery organization of the network. Our near-exact predictions highlight how accurately capturing the long-range correlations in network structures allows to easily and effectively compress real complex network data.

A general form of smart contract for decentralized energy systems management

Smart contract platforms have the potential to allow shared automatic control of energy transfer within networks in a replicable, secure, verifiable and trustworthy way. Here we present a general form of smart contract which captures the elements needed for shared control that will help formalize decentralization. Two mechanisms were defined for agreement of control instructions for a medium-voltage direct-current (MVDC) link connecting two separately operated 33 kV distribution networks. These were instantiated as smart contracts and were evaluated in terms of cost and the computational requirements for their execution. Real network and converter data from the ANGLE-DC demonstration project were used to model the MVDC link. We demonstrate that using smart contracts to agree control instructions between different parties is feasible. The potential for shared control using smart contracts gives operators and regulators a way of defining and decentralizing operating responsibilities within energy systems.

Modeling information popularity dynamics based on branching process

In the age of Web 2.0, modeling and predicting the popularity of online information was an important issue in information dissemination. Online social medium greatly affects the way we communicate with each other. However, little is known about what fundamental mechanisms drive the dynamical information flow in online social systems. To address this problem, we develop a theoretical probabilistic model based on branching process to characterize the process in which micro-blog information gains its popularity. Firstly, the data of information popularity and network structure of micro-blog network are analyzed. The statistical results show that the attenuation of information popularity follows a scaling law whose exponent is 1.8, and in-degree and out-degree of micro-blog network each also obey a power law distribution whose exponent is 1.5. The results of power law distribution show that there is a high-degree heterogeneity in a micro-blog system. The proportion of micro-blog information with popularity less than 100 is 95.8%, while the amount of micro-blog information with popularity more than 10, 000 is very small. The number of fans (in-degree) less than 100 accounts for 56.4%, while some users have millions of fans.Secondly, according to the design mechanism of the Weibo system, we assume that each user has two lists, i.e. a "home page list" and a "personal page list". Meanwhile, each user has two states at each moment: generating a new message with probability <inline-formula><tex-math id="Z-20190326105842-2">\begin{document}${\mu} $\end{document}</tex-math><alternatives><graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="7-20181948_Z-20190326105842-2.jpg"/><graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="7-20181948_Z-20190326105842-2.png"/></alternatives></inline-formula> to be sent out; 2) or forwarding the information already on the "personal page list" with probability <inline-formula><tex-math id="Z-20190326105842-3">\begin{document}$ (1-{\mu}) $\end{document}</tex-math><alternatives><graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="7-20181948_Z-20190326105842-3.jpg"/><graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="7-20181948_Z-20190326105842-3.png"/></alternatives></inline-formula> . Based on the assumptions, the information popularity model is proposed. Finally, the model is simulated. The simulation results show that the model can reproduce some features of real social network data, and the popularity of information is related to the network structure. By solving the model equation, the results of theoretical prediction are consistent with the simulation analyses and actual data.

Multi-Layer Network Local Community Detection Based on Influence Relation

In recent years, the discovery of local communities in multi-layer networks has become an active research field of complex systems. Such as communication, social networking, sensor network the rapid development of new technologies, the amount of data generated by increased, all want to obtain the network information difficulty is big, and the network, community, mutual influence between nodes, greatly enhance the complexity of network and make local found existing multilayer network method is unable to get more accurate test results. In this paper, based on the homogeneity drive of multi-layer network and the influence relation index of multi-layer path length measurement, a local community detection model based on the influence relation of the multi-layer network is proposed by combining the direct influence relation and indirect influence relation of the network (IMLC). Compared with six real multi-layer network data sets, the algorithm has better robustness in many most advanced multi-layer methods: GL, PMM, and ML-LCD.

Research on collaborative filtering recommendation algorithm based on social network

For users of social-based social networking services, we propose a local random walk-based friend recommendation approach by bringing together social network and tie strength. We firstly construct a weighted friend network as the basis for friend recommendation. Then, users' similarity is determined by a local random walk-based similarity measure on a weighted friend network. Experiments show that we use real social network data to evaluate the new method. The validity of the method is illustrated.