Skip to main content
main-content
Top

About this book

I was invited to join the Organizing Committee of the First International Conference on Complex Sciences: Theory and Applications (Complex 2009) as its ninth member. At that moment, eight distinguished colleagues, General Co-chairs Eugene Stanley and Gaoxi Xiao, Technical Co-chairs János Kertész and Bing-Hong Wang, Local Co-chairs Hengshan Wang and Hong-An Che, Publicity Team Shi Xiao and Yubo Wang, had spent hundreds of hours pushing the conference half way to its birth. Ever since then, I have been amazed to see hundreds of papers flooding in, reviewed and commented on by the TPC members. Finally, more than 200 contributions were - lected for the proceedings currently in your hands. They include about 200 papers from the main conference (selected from more than 320 submissions) and about 33 papers from the five collated workshops: Complexity Theory of Art and Music (COART) Causality in Complex Systems (ComplexCCS) Complex Engineering Networks (ComplexEN) Modeling and Analysis of Human Dynamics (MANDYN) Social Physics and its Applications (SPA) Complex sciences are expanding their colonies at such a dazzling speed that it - comes literally impossible for any conference to cover all the frontiers.

Table of Contents

Frontmatter

Part I

Frontmatter

Return Intervals Approach to Financial Fluctuations

Financial fluctuations play a key role for financial markets studies. A new approach focusing on properties of return intervals can help to get better understanding of the fluctuations. A return interval is defined as the time between two successive volatilities above a given threshold. We review recent studies and analyze the 1000 most traded stocks in the US stock markets. We find that the distribution of the return intervals has a well approximated scaling over a wide range of thresholds. The scaling is also valid for various time windows from one minute up to one trading day. Moreover, these results are universal for stocks of different countries, commodities, interest rates as well as currencies. Further analysis shows some systematic deviations from a scaling law, which are due to the nonlinear correlations in the volatility sequence. We also examine the memory in return intervals for different time scales, which are related to the long-term correlations in the volatility. Furthermore, we test two popular models, FIGARCH and fractional Brownian motion (fBm). Both models can catch the memory effect but only fBm shows a good scaling in the return interval distribution.

Fengzhong Wang, Kazuko Yamasaki, Shlomo Havlin, H. Eugene Stanley

Organizational Adaptative Behavior: The Complex Perspective of Individuals-Tasks Interaction

Organizations with different organizational structures have different organizational behaviors when responding environmental changes. In this paper, we use a computational model to examine organizational adaptation on four dimensions: Agility, Robustness, Resilience, and Survivability. We analyze the dynamics of organizational adaptation by a simulation study from a complex perspective of the interaction between tasks and individuals in a sales enterprise. The simulation studies in different scenarios show that more flexible communication between employees and less hierarchy level with the suitable centralization can improve organizational adaptation.

Jiang Wu, Duoyong Sun, Bin Hu, Yu Zhang

Optimization Using a New Bio-inspired Approach

There is growing interest in bio(logy)-inspired approaches that are inspired by the principles of biology and that can solve difficult problems. In this paper, we propose a new computational algorithm that is inspired by molecular mechanics for the solution of complex problems. There is a deep and useful connection between mechanics mechanics and combinatorial optimization. This connection exposes new information and allows an unfamiliar perspective on traditional optimization problems and approaches. The alternative of

molecular mechanics algorithm

(MMA) to traditional approaches has the advantages of inherent parallelism and the ability to deal with a variety of complicated social interactions, autonomous behaviors and multiple objectives.

Xiang Feng, Francis C. M. Lau, Daqi Gao

Optimality Conditions of a Three-Dimension Non-smooth Thermodynamic System of Sea Ice

This study is intended to provide the mathematical foundation for the numerical computation of the parameter identification problems of the three-dimensional two-layer thermodynamic system of sea ice. The non-smooth thermodynamic system with mixed boundary conditions is established, its properties are obtained and the first-order necessary conditions of the parameter identification problem of the non-smooth system are derived.

Wei Lv, Hong Bao, Enmin Feng

Optimal Service Capacities in a Competitive Multiple-Server Queueing Environment

The study of economic behavior of service providers in a competition environment is an important and interesting research issue. A two-server queueing model has been proposed in Kalai et al. [11] for this purpose. Their model aims at studying the role and impact of service capacity in capturing larger market share so as to maximize the long-run expected profit. They formulate the problem as a two-person strategic game and analyze the equilibrium solutions. The main aim of this paper is to extend the results of the two-server queueing model in [11] to the case of multiple servers. We will only focus on the case when the queueing system is stable.

Wai-Ki Ching, Sin-Man Choi, Min Huang

One Kind of Network Complexity Pyramid with Universality and Diversity

It is based on well-known network models Euler graph, Erdös and Renyi random graph, Watts-Strogatz small-world model and Barabási-Albert scale-free networks, and combined the unified hybrid network theoretical frame. One kind of network complexity pyramid with universality and diversity is constructed, described and reviewed. It is found that most unweighted and weighted models of network science can be investigated in a unification form using four hybrid ratios (

dr,fd,gr,vg

). As a number of hybrid ratios increase, from the top level to the bottom level complexity and diversity of the pyramid is increasing but universality and simplicity is decreasing. The network complexity pyramid may have preferable understanding in complicated transition relationship between complexity-diversity and simplicity-universality.

Jin-Qing Fang, Yong Li

On Traveling Diameter of an Instance of Complex Networks – Internet

As an instance of complex networks, Internet has been a hot topic for both complex networks and traditional networks research fields. Internet Traveling Diameter (ITD) is an important property defined in this paper representing the dynamic flow of Internet performance, and was mainly discussed. Short-term forecast model of ITD was firstly studied, then after it was proved that the short-term one was not good enough for long term forecast due to the complexity of Internet, the long-term model was studied. Both short-term and long-term model were given their mathematical descriptions at last.

Ye Xu, Zhuo Wang, Wen-bo Zhang

On the Approximation Solution of a Cellular Automaton Traffic Flow Model and Its Relationship with Synchronized Flow

This paper studies approximation solution of a cellular automaton model. In the model, the finite size effect is trivial because the congested flow is quite homogeneous. Thus, the approximation solution of a small sized system can be regarded as solution of large system. We have investigated the approximation solution of a small traffic system with two vehicles. The analytical result is in good agreement with simulation. Finally, it is demonstrated that the homogeneous congested flow is closely related to synchronized flow.

R. Jiang, Y. M. Yuan, K. Nishinari

On Scale-Free Prior Distributions and Their Applicability in Large-Scale Network Inference with Gaussian Graphical Models

This paper concerns the specification, and performance, of scale-free prior distributions with a view toward large-scale network inference from small-sample data sets. We devise three scale-free priors and implement them in the framework of Gaussian graphical models. Gaussian graphical models are used in gene network inference where high-throughput data describing a large number of variables with comparatively few samples are frequently analyzed by practitioners. And, although there is a consensus that many such networks are scale-free, the

modus operandi

is to assign a random network prior. Simulations demonstrate that the scale-free priors outperform the random network prior at recovering scale-free trees with degree exponents near 2, such as are characteristic of many real-world systems. On the other hand, the random network prior compares favorably at recovering scale-free trees characterized by larger degree exponents.

Paul Sheridan, Takeshi Kamimura, Hidetoshi Shimodaira

On General Laws of Complex Networks

By introducing and analyzing a renormalization procedure, Song et al. [1] draw the conclusion that many complex networks exhibit self-repeating patterns on all length scales. First, we aim to demonstrate that the aforementioned conclusion is inadequately justified, mainly because their equation (7) on the invariance of degree distribution under renormalization does not hold in general. Secondly, Barabási and Albert [2] find that many large networks exhibit a scale-free power-law distribution of vertex degrees. They show this common feature to be a consequence of two generic mechanisms: (i) networks expand continuously by the addition of new vertices, and (ii) new vertices attach preferentially to those that are already well connected. We show that when vertex degrees of large networks follow a scale-free power-law distribution with the exponent

γ

 ≥ 2, the number of degree-1 vertices, when nonzero, is of the same order as the network size N and that the average degree is of order less than log N. Given that many real networks satisfy these two conditions, our results add another necessary characteristic of the scale-free power-law distribution of vertex degrees in such networks. Our method has the benefit of relying on conditions that are static and easily verified. They are verified by many experimental results of diverse real networks.

Wenjun Xiao, Limin Peng, Behrooz Parhami

On Distributed Multi-Point Concurrent Test System and Its Implementation

As the rapid expansion of the Internet, novel network applications are constantly emerging accompanied by the increasingly growing of the complexity for network protocols. As a result, IP routers are becoming more and more important in today’s Internet. However, test methodology and test systems for IP routers often fall behind the state of the art. Based on the deficiency of current stand-alone test systems, this paper analyzes the necessity for distributed test architecture and introduces a new test system called Distributed Multi-point Concurrent Test System (DMC-TS) which can mirror real-world networks in the test experiments and conduct conformance testing, performance testing and interoperability testing for the routers under test. Key issues for the implementation of DMC-TS are discussed, especially on test synchronization problem. Consequently, two types of test synchronization problem are pointed out with the corresponding solutions. Some experiments are carried out to illustrate the feasibility and practicability of DMC-TS.

Hao Luo, Huaxin Zeng

Organizational Structure of the Transcriptional Regulatory Network of Yeast: Periodic Genes

In this paper we investigate the organizational structure of the transcriptional regulatory network of

S. cerevisiae

with respect to the connectivity structure of periodic genes. We demonstrate that the giant strongly connected component plays a prominent role serving as central connector for genes experimentally found to be periodically expressed during the cell cycle of yeast. Numerically, we find by randomization of the gene labels that this organizational structure is unlikely to be formed by chance.

Frank Emmert-Streib, Matthias Dehmer

Packet-Level Traffic Allocation for Real-Time Streaming over Multipath Networks

We address packet-level traffic allocation problem for real-time media streaming under multipath network environment. Based on an in-depth analysis of multipath real-time streaming model, also considering fluctuation of multipath network status as well as burst of media sending rate, we suggest that traffic load should be allocated to paths in proportion to the paths’ available bandwidths, which minimizes the overall bandwidth overload probability. Moreover, due to the smallest transmission unit is packet, in order to execute the traffic allocation policy exactly, weighted size-aware packet distribution algorithm is proposed to avoid the actual traffic deviation due to variance of packet sizes. Simulation results show that the proposed algorithm outperforms other traditional algorithms, especially for reducing packet late arrivals, which has negative impaction in real-time transmission.

Yanfeng Zhang, Cuirong Wang, Yuan Gao

Particle Competition in Complex Networks for Semi-supervised Classification

Semi-supervised learning is an important topic in machine learning. In this paper, a network-based semi-supervised classification method is proposed. Class labels are propagated by combined random-deterministic walking of particles and competition among them. Different from other graph-based methods, our model does not rely on loss function or regularizer. Computer simulations were performed with synthetic and real data, which show that the proposed method can classify arbitrarily distributed data, including linear non-separable data. Moreover, it is much faster due to lower order of complexity and it can achieve better results with few pre-labeled data than other graph based methods.

Fabricio Breve, Liang Zhao, Marcos Quiles

Retail Location Choice with Complementary Goods: An Agent-Based Model

This paper models the emergence of retail clusters on a supply chain network comprised of suppliers, retailers, and consumers. Firstly, an agent-based model is proposed to investigate retail location distribution in a market of two complementary goods. The methodology controls for supplier locales and unit sales prices of retailers and suppliers, and a consumer’s willingness to patronize a retailer depends on the total travel distance of buying both goods. On a circle comprised of discrete locations, retailers play a non-cooperative game of location choice to maximize individual profits. Our findings suggest that the probability distribution of the number of clusters in equilibrium follows power law and that hierarchical distribution patterns are much more likely to occur than the spread-out ones. In addition, retailers of complementary goods tend to co-locate at supplier locales. Sensitivity tests on the number of retailers are also performed. Secondly, based on the County Business Patterns (CBP) data of Minneapolis-St. Paul from US Census 2000 database, we find that the number of clothing stores and the distribution of food stores at the zip code level follows power-law distribution.

Arthur Huang, David Levinson

Research on Web2.0 System Design Based on CAS Theory

According to complexity theory, this paper analyses several characteristics of some present Web2.0 systems, such as Blog, Wiki, SNS and social tags. It also summarizes the disadvantages of current information system design methods and finally re-designs it based on CAS and DSDM.

Kai Chen, Hengshan Wang

Reconstructing Gene Networks from Microarray Time-Series Data via Granger Causality

Reconstructing gene network structure from Microarray time-series data is a basic problem in Systems Biology. In gene regulation networks, the time delays and the combination effects which are not considered by most existent models are key factors to understand the genetic regulatory networks. To address these problems, this paper proposed a fast algorithm to learn initial network structures for gene networks from time-series data by employing the Granger causality model to analyze the time delays and the combination effects for gene regulation. The simulation results on a synthetic network and the ethylene pathway in

Arabidopsis

show that the proposed algorithm is a promise tool for learning network structures from time-series data.

Qiang Luo, Xu Liu, Dongyun Yi

Recognition of Important Subgraphs in Collaboration Networks

We propose a method for recognition of most important subgraphs in collaboration networks. The networks can be described by bipartite graphs, where basic elements, named actors, are taking part in events, organizations or activities, named acts. It is suggested that the subgraphs can be described by so-called

k

-cliques, which are defined as complete subgraphs of two or more vertices. The

k

-clique act degree is defined as the number of acts, in which a

k

-clique takes part. The

k

-clique act degree distribution in collaboration networks is investigated via a simplified model. The analytic treatment on the model leads to a conclusion that the distribution obeys a so-called shifted power law

P

(

q

) ∝ (

q

 + 

α

)

 − 

γ

where

α

and

γ

are constants. This is a very uneven distribution. Numerical simulations have been performed, which show that the model analytic conclusion remains qualitatively correct when the model is revised to approach the real world evolution situation. Some empirical investigation results are presented, which support the model conclusion. We consider the cliques, which take part in the largest number of acts, as the most important ones. With this understanding we are able to distinguish some most important cliques in the real world networks.

Chun-Hua Fu, Yue-Ping Zhou, Xiu-Lian Xu, Hui Chang, Ai-Xia Feng, Jian-Jun Shi, Da-Ren He

Queueing Transition of Directed Polymer in Random Media with a Defect

We study a queueing transition of directed polymer in random media with an attractive defect at the center of the one dimensional substrate. The end to end distance

Δx

of the polymer follows

Δx

~

t

1/

z

with

z

 = 3/2, for weak defect strength

ε

where

t

is the polymer length. If

ε

 ≥ 

ε

c

then the polymer is localized with finite

Δx

in long

t

limit. The transition is related to the queueing phenomena of the asymmetric simple exclusion process.

Jae Hwan Lee, Jin Min Kim

Pollution Modeling and Simulation with Multi-Agent and Pretopology

Pollution in metropolitan cities has become a serious problem, resulting in poor living conditions and serious health problems. Pollution being qualified as a complex system, we propose a multi-agent approach to model and simulate it, so that we could study, analyze and predict it better. As in the early stage of the project, we have some successful experiments and attempts to integrate the mathematical theory of Pretopology in the modeling and simulation levels. In addition, these interesting results shades some light on our future direction.

Murat Ahat, Sofiane Ben Amor, Marc Bui, Michel Lamure, Marie-Françoise Courel

Policy, Design and Management: The in-vivo Laboratory for the Science of Complex System

Complex systems scientists cannot by themselves perform experiments on complex socio-technical systems. The best they can do is to perform experiments alongside policy makers who are constantly engaged in experiments as they design and manage the systems the systems for which they are responsible. In this context the nature of

prediction

in the implementation of real systems is much more complicated than it is in traditional science. The

goals

identified by policymakers change through time, and this is usually managed through the design and management processes. The combination of policy and design is the opportunity - the only opportunity - for complex systems scientists to engage and to be allowed to be involved in

in-vivo

experiments in large socio-technical systems. In turn this opens up new methodological approaches and questions for the science of complex systems.

Jeffrey Johnson

Phase Transition of Active Rotators in Complex Networks

We study the nonequilibrium phenomena of a coupled active rotator model in complex networks. From a numerical Langevin simulation, we find the peculiar phase transition not only on globally connected network but also on other complex networks and reveal the corresponding phase diagram. In this model, two phases — stationary and quasi-periodic moving phases — are observed, in which microscopic dynamics are thoroughly investigated. We extend our study to the non-identical oscillators and the more heterogeneous degree distribution of complex networks.

Seung-Woo Son, Hawoong Jeong, Hyunsuk Hong

Personal Recommendation in User-Object Networks

Thanks to the Internet and the World Wide Web, we live in a world of many possibilities we can choose from thousands of movies, millions of books, and billions of web pages. Far exceeding our personal processing capacity, this excessive freedom of choice calls for automated ways to find the relevant information. As a result, the field of information filtering is very active and rich with unanswered challenges. In this short paper, I will give a brief introduction on the design of recommender systems, which recommend objects to users based on the historical records of users’ activities. A diffusion-based recommendation algorithm, as well as two improved algorithms are investigated. Numerical results on a benchmark data set have demonstrated the advantages in algorithmic accuracy.

Tao Zhou

Performance Analysis of Public Transport Systems in Nanjing Based on Network Topology

The urban public transport network (UPTN) in Nanjing is characterized by a complex network with topological pedestals. The empirical data indicates that it is a small-world network. Under malicious attack to the high connectivity nodes of the network, the average path-length will increase 2.5 times, the reliability and traffic capacity of the UPTN will greatly decline, and the travel expenditure will distinctively increase. The topological significance of stations and routes are redefined to help assess the small-world property of UPTNs, so as to improve city transportation. It is also found that if the urban rail transit, such as metro, is introduced to the UPTN, then the topological diameter of the network is reduced, and its structure is optimized.

Ping Li, Zhen-Tao Zhu, Jing Zhou, Jin-Yuan Ding, Hong-Wei Wang, Shan-Sen Wei

Non-sufficient Memories That Are Sufficient for Prediction

The causal states of computational mechanics define the minimal sufficient (prescient) memory for a given stationary stochastic process. They induce the

ε

-machine which is a hidden Markov model (HMM) generating the process. The

ε

-machine is, however, not the minimal generative HMM and minimal internal state entropy of a generative HMM is a tighter upper bound for excess entropy than provided by statistical complexity. We propose a notion of prediction that does not require sufficiency. The corresponding models can be substantially smaller than the

ε

-machine and are closely related to generative HMMs.

Wolfgang Löhr, Nihat Ay

New Statistics for Testing Differential Expression of Pathways from Microarray Data

Exploring biological meaning from microarray data is very important but remains a great challenge. Here, we developed three new statistics: linear combination test, quadratic test and de-correlation test to identify differentially expressed pathways from gene expression profile. We apply our statistics to two rheumatoid arthritis datasets. Notably, our results reveal three significant pathways and 275 genes in common in two datasets. The pathways we found are meaningful to uncover the disease mechanisms of rheumatoid arthritis, which implies that our statistics are a powerful tool in functional analysis of gene expression data.

Hoicheong Siu, Hua Dong, Li Jin, Momiao Xiong

Multiple Phase Transitions in the Culture Dissemination

We study the coevolution process in the Axelrod’s model with the consideration of agents’ abilities to access to the information. With a parameter to control the ability of communication, we observe two kinds of phase transitions both for cultural domains and network fragments, respectively. With the simulation results, we find the relationship between the critical value and the controlled parameter. The results indicate that the powerful ability to access to the information benefits the dissemination of culture in the system.

Bing Wang, Yuexing Han, Luonan Chen, Kazuyuki Aihara

Joint Channel-Network Coding (JCNC) for Distributed Storage in Wireless Network

We propose to construct a joint channel-network coding (knosswn as Random Linear Coding) scheme based on improved turbo codes for the distributed storage in wireless communication network with k data nodes, s storage nodes (k<s) and a data collector. This framework extends the classical distributed storage with erasure channel to AWGN and fading channel scenario. We investigate the throughput performance of the Joint Channel-Network Coding (JCNC) system benefits from network coding, compared with that of system without network coding based only on store and forward (S-F) approach. Another helpful parameter: node degree (L) indicates how many storage nodes one data packet should fall onto. L characterizes the en/decoding complexity of the system. Moreover, this proposed framework can be extended to ad-hoc and sensor network easily.

Ning Wang, Jiaru Lin

Invariance of the Hybrid System in Microbial Fermentation

In this study, we propose a nonlinear hybrid dynamical system to describe the concentrations of extracellular and intracellular substances in the process of bio-dissimilation of glycerol to 1,3-propanediol. An invariance principle is established for the hybrid dynamical system. At the same time, we state and prove new stability criteria for the nonlinear hybrid system. These results provide less conservative stability conditions for hybrid system as compared to classical results in the literature and allow us to characterize the invariance of a class of nonlinear hybrid dynamical systems.

Caixia Gao, Enmin Feng

Is Self-organization a Rational Expectation?

A Critical Review of Complexity and Emergence

Over decades and under varying names the study of biology-inspired algorithms applied to non-living systems has been the subject of a small and somewhat exotic research community. Only the recent coincidence of a growing inability to master the design, development and operation of increasingly intertwined systems and processes, and an accelerated trend towards a naïve if not romanticizing view of nature in the sciences, has led to the adoption of biology-inspired algorithmic research by a wider range of sciences. Adaptive systems, as we apparently observe in nature, are meanwhile viewed as a promising way out of the complexity trap and, propelled by a long list of ‘self’ catchwords, complexity research has become an influential stream in the science community. This paper presents four provocative theses that cast doubt on the strategic potential of complexity research and the viability of large scale deployment of biology-inspired algorithms in an expectation driven world.

Heinz Luediger

Inter-Profile Similarity (IPS): A Method for Semantic Analysis of Online Social Networks

Online Social Networks (OSN)[OSN] are experiencing an explosive growth rate and are becoming an increasingly important part of people’s lives. There is an increasing desire to aid online users in identifying potential friends, interesting groups, and compelling products to users. These networks have offered researchers almost total access to large corpora of data. An interesting goal in utilizing this data is to analyze user profiles and identify how similar subsets of users are. The current techniques for comparing users are limited as they require common terms to be shared by users. We present a simple and novel extension to a word-comparison algorithm [6], entitled Inter-Profile Similarity (IPS), which allows comparison of short text phrases

even if they share no common terms

. The output of Inter-Profile Similarity (IPS) is simply a scalar value in [0,1], with 1 denoting complete similarity and 0 the opposite. Therefore it is easy to understand and can provide a total ordering of users. We, first, evaluated the effectiveness of Inter-Profile Similarity (IPS) with a user-study, and then applied it to datasets from Facebook and Orkut verifying and extending earlier results. We show that Inter-Profile Similarity (IPS) yields both a larger range for the similarity value and obtains a higher value than intersection-based mechanisms. Both Inter-Profile Similarity (IPS) and the output from the analysis of the two Online Social Networks (OSN)[OSN] should help to predict and classify social links, make recommendations, and annotate friends relations for social network analysis.

Matt Spear, Xiaoming Lu, Norman S. Matloff, S. Felix Wu

Inefficiency in Networks with Multiple Sources and Sinks

We study the problem of optimizing traffic in decentralized transportation networks, where the cost of a link depends on its congestion. If users of a transportation network are permitted to choose their own routes, they generally try to minimize their personal travel time. In the absence of centralized coordination, such a behavior can be inefficient for society and even for each individual user. This inefficiency can be quantified by the “price of anarchy”, the ratio of the suboptimal total cost to the socially optimal cost. Here we study the price of anarchy in multi-commodity networks, (i.e., networks where traffic simultaneously flows between different origins and destinations).

Hyejin Youn, Michael T. Gastner, Hawoong Jeong

Impacts of Local Events on Communities and Diseases

The study of community networks has attracted considerable attention recently. In this paper, we propose an evolving community network model based on local events, the addition of new nodes intra-community and new links intra- or inter-community. Employing growth and preferential attachment mechanisms, we generate the network with a generalized power-law distribution of nodes’ degrees. Furthermore, we study epidemic spreading in the resulting network by the simple SIS model to understand the influence of the network structure on the dynamics. We find that the existence of communities in networks causes the critical behavior of the spreading dynamics and keeps epidemics endemic.

Xin-Jian Xu, Li-Jie Zhang, Guo-Hong Yang, Xun Zhang

Identifying Social Communities in Complex Communications for Network Efficiency

Complex communication networks, more particular Mobile Ad Hoc Networks (MANET) and Pocket Switched Networks (PSN), rely on short range radio and device mobility to transfer data across the network. These kind of mobile networks contain duality in nature: they are radio networks at the same time also human networks, and hence knowledge from social networks can be also applicable here. In this paper, we demonstrate how identifying social communities can significantly improve the forwarding efficiencies in term of delivery ratio and delivery cost. We verify our hypothesis using data from five human mobility experiments and test on two application scenarios, asynchronous messaging and publish/subscribe service.

Pan Hui, Eiko Yoneki, Jon Crowcroft, Shu-Yan Chan

Hypernetworks of Complex Systems

Hypernetworks generalise the concept of a relation between two things to relations between many things. The notion of

relational simplex

generalises the concept of network edge to relations between many elements. Relational simplices have multi-dimensional connectivity related to hyper-graphs and the Galois lattice of maximally connected sets of elements. This structure acts as a kind of

backcloth

for the dynamic system

traffic

represented by numerical mappings, where the topology of the backcloth constrains the dynamics of the traffic. Simplices provide a way of defining multilevel structure. This relates to system time measured by the formation of simplices as system events. Multilevel hypernetworks are classes of sets of relational simplices that represent the system backcloth and the traffic of systems activity it supports. Hypernetworks provide a significant generalisation of network theory, enabling the integration of relational structure, logic, and topological and analytic dynamics. They provide structures that are likely to be necessary if not sufficient for a science of complex multilevel socio-technical systems.

Jeffrey Johnson

Less Restrictive Synchronization Criteria in Complex Networks with Coupling Delays

This paper considers the synchronization in complex networks with coupling delays, whose topologies could be be symmetric and asymmetric. Differing from most works on the synchronization in complex networks with coupling delays, this paper only uses a positively-defined function, which is definitely not a Krasovskii-Lyapunov function, to analyze the synchronization criteria. Further, we can derive novel but less restrictive synchronization criteria than those resulting from the Krasovskii-Lyapunov theory. Theoretical analysis and numerical simulations fully verify the main results.

Yun Shang, Maoyin Chen

MANIA: A Gene Network Reverse Algorithm for Compounds Mode-of-Action and Genes Interactions Inference

Understanding the complexity of the cellular machinery represents a grand challenge in molecular biology. To contribute to the deconvolution of this complexity, a novel inference algorithm based on linear ordinary differential equations is proposed, based on high-throughput gene expression data. The algorithm can infer (i) gene-gene interactions from steady state expression profiles

AND

(ii) mode-of-action of the components that can trigger changes in the system. Results demonstrate that the proposed algorithm can identify

both

information with high performances, thus overcoming the limitation of current algorithms that can infer reliably only one.

Darong Lai, Hongtao Lu, Mario Lauria, Diego di Bernardo, Christine Nardini

Measurement and Statistics of Application Business in Complex Internet

Owing to independent topologies and autonomic routing mechanism, the logical networks formed by Internet application business behavior cause the significant influence on the physical networks. In this paper, the backbone traffic of TUNET (Tsinghua University Networks) is measured, further more, the two most important application business: HTTP and P2P are analyzed at IP-packet level. It is shown that uplink HTTP and P2P packets behavior presents spatio-temporal power-law characteristics with exponents 1.25 and 1.53 respectively. Downlink HTTP packets behavior also presents power-law characteristics, but has more little exponents

γ

= 0.82 which differs from traditional complex networks research result. Moreover, downlink P2P packets distribution presents an approximate power-law which means that flow equilibrium profits little from distributed peer-to peer mechanism actually.

Lei Wang, Yang Li, Yipeng Li, Shuhang Wu, Shiji Song, Yong Ren

Moving Breather Collisions in the Peyrard-Bishop DNA Model

We consider collisions of moving breathers (MBs) in the Peyrard-Bishop DNA model. Two identical stationary breathers, separated by a fixed number of pair-bases, are perturbed and begin to move approaching to each other with the same module of velocity. The outcome is strongly dependent of both the velocity of the MBs and the number of pair-bases that initially separates the stationary breathers. Some collisions result in the generation of a new stationary trapped breather of larger energy. Other collisions result in the generation of two new MBs. In the DNA molecule, the trapping phenomenon could be part of the complex mechanisms involved in the initiation of the transcription processes.

A. Alvarez, F. R. Romero, J. Cuevas, J. F. R. Archilla

Morphological Similarities between DBM and an Economic Geography Model of City Growth

An urban microeconomic model of households evolving in a 2D cellular automata allows to simulate the growth of a metropolitan area where land is devoted to housing, road network and agricultural/green areas. This system is self-organised: based on individualistic decisions of economic agents who compete on the land market, the model generates a metropolitan area with houses, roads, and agriculture. Several simulation are performed. The results show strong similarities with physical Dieletric breackdown models (DBM). In particular, phase transitions in the urban morphology occur when a control parameter reaches critical values. Population density in our model and the electric potential in DBM play similar roles, which can explain these resemblances.

Jean Cavailhès, Pierre Frankhauser, Geoffrey Caruso, Dominique Peesters, Isabelle Thomas, Gilles Vuidel

Modular Synchronization in Complex Network with a Gauge Kuramoto Model

We modify the Kuramoto equation(KE) by introducing a gauge term which is a function of link betweenness centrality(BC). The gauge term induces the phase difference from 0 to

π

between two nodes that belong to different modules. Therefore, a synchronization occurs in each module individually even though the whole network is not synchronized globally. By measuring the phase similarity of all pairs of connected nodes, we can detect the modular structure of complex networks. This algorithm requires relatively little computational time

O

(

NL

) for network with

N

nodes and

L

links.

Chulho Choi, Eulsik Oh, Byungnam Kahng, Doochul Kim

Modification Propagation in Complex Networks

To keep up with rapidly changing conditions, business systems and their associated networks are growing increasingly intricate as never before. By doing this, network management and operation costs not only rise, but are difficult even to measure. This fact must be regarded as a major constraint to system optimization initiatives, as well as a setback to derived economic benefits. In this work we introduce a simple model in order to estimate the relative cost associated to modification propagation in complex architectures. Our model can be used to anticipate costs caused by network evolution, as well as for planning and evaluating future architecture development while providing benefit optimization.

Mary Luz Mouronte, María Luisa Vargas, Luis Gregorio Moyano, Francisco Javier García Algarra, Luis Salvador Del Pozo

Modelling of Population Migration to Reproduce Rank-Size Distribution of Cities in Japan

We investigate the rank-size distribution of cities in Japan by data analysis and computer simulation. From our previous data analysis of the census data after World War II, it has been clarified that the power exponent of the rank-size distribution of cities changes with time and Zipf’s law holds only for a restricted period. We show that Zipf’s law broke down owing to the great mergers and recovered by investigating the time evolution of the rank-size distribution of cities without mergers.

Hiroto Kuninaka, Mitsugu Matsushita

Modeling and Robustness Analysis of Biochemical Networks of Glycerol Metabolism by Klebsiella Pneumoniae

Glycerol bioconversion to 1,3-propanediol (1,3-PD) by

Klebsiella pneumoniae

(

K. pneumoniae

) can be characterized by an intricate network of interactions among biochemical fluxes, metabolic compounds, key enzymes and genetic regulatory. To date, there still exist some uncertain factors in this complex network because of the limitation in bio-techniques, especially in measuring techniques for intracellular substances. In this paper, among these uncertain factors, we aim to infer the transport mechanisms of glycerol and 1,3-PD across the cell membrane, which have received intensive interest in recent years. On the basis of different inferences of the transport mechanisms, we reconstruct various metabolic networks correspondingly and subsequently develop their dynamical systems (S-systems). To determine the most reasonable metabolic network from all possible ones, we establish a quantitative definition of biological robustness and undertake parameter identification and robustness analysis for each system. Numerical results show that it is most possible that both glycerol and 1,3-PD pass the cell membrane by active transport and passive diffusion.

Jianxiong Ye, Enmin Feng, Lei Wang, Zhilong Xiu, Yaqin Sun

Modeling and Properties of Nonlinear Stochastic Dynamical System of Continuous Culture

The stochastic counterpart to the deterministic description of continuous fermentation with ordinary differential equation is investigated in the process of glycerol bio-dissimilation to 1,3-propanediol by

Klebsiella pneumoniae

. We briefly discuss the continuous fermentation process driven by three-dimensional Brownian motion and Lipschitz coefficients, which is suitable for the factual fermentation. Subsequently, we study the existence and uniqueness of solutions for the stochastic system as well as the boundedness of the Two-order Moment and the Markov property of the solution. Finally stochastic simulation is carried out under the Stochastic Euler-Maruyama method.

Lei Wang, Enmin Feng, Jianxiong Ye, Zhilong Xiu

Modeling a Complex Biological Network with Temporal Heterogeneity: Cardiac Myocyte Plasticity as a Case Study

Complex biological systems often characterize nonlinear dynamics. Employing traditional deterministic or stochastic approaches to quantify these dynamics either fail to capture their existing deviant effects or lead to combinatorial explosion. In this work we devised a novel approach that projects the biological functions within a pathway to a network of stochastic events that are random in time and space. By applying this approach recursively to the object system we build the event network of the entire system. The dynamics of the system evolves through the execution of the event network by a simulation engine which comprised of a time prioritized event queue. As a case study we utilized the current method and conducted an in-silico experiment on the metabolic plasticity of a cardiac myocyete. We aimed to quantify the down stream effects of insulin signaling that predominantly controls the plasticity in myocardium. Intriguingly, our in-silico results on transcription regulatory effect of insulin showed a good agreement with experimental data. Meanwhile we were able to characterize the flux change across major metabolic pathways over 48 hours of the in-silico experiment. Our simulation performed a remarkable efficiency by conducting 48 hours of simulation-time in less that 2 hours of processor time.

Amin R. Mazloom, Kalyan Basu, Subhrangsu S. Mandal, Sajal K. Das

Model and Dynamic Behavior of Malware Propagation over Wireless Sensor Networks

Based on the inherent characteristics of wireless sensor networks (WSN), the dynamic behavior of malware propagation in flat WSN is analyzed and investigated. A new model is proposed using 2-D cellular automata (CA), which extends the traditional definition of CA and establishes whole transition rules for malware propagation in WSN. Meanwhile, the validations of the model are proved through theoretical analysis and simulations. The theoretical analysis yields closed-form expressions which show good agreement with the simulation results of the proposed model. It is shown that the malware propaga-tion in WSN unfolds neighborhood saturation, which dominates the effects of increasing infectivity and limits the spread of the malware. MAC mechanism of wireless sensor networks greatly slows down the speed of malware propagation and reduces the risk of large-scale malware prevalence in these networks. The proposed model can describe accurately the dynamic behavior of malware propagation over WSN, which can be applied in developing robust and efficient defense system on WSN.

Yurong Song, Guo-Ping Jiang

Measuring the Efficiency of Network Designing

Network designing often involves two significant yet contradictive objectives: enhancing the whole network’s transmission efficiency while at the same time lowering the whole network’s designing cost. Deep study of the interplay between major aspects of network planning– network topology, routing algorithm and node’s transmission capability configuration–reveals that good tradeoff can be achieved between these two objectives. By properly combining network topology, routing algorithm and node capability configuration scheme, the network can achieve desirable transmission efficiency at very low cost. This discovery will undoubtedly provide insight into the next generation data network designing.

Guoqiang Zhang, Guoqing Zhang

Gravity Model for Transportation Network Based on Optimal Expected Traffic

We propose a spatial network model for transportation system based on the optimal expected traffic. The expected traffic represents the prediction of the flow created by two vertices and is calculated by the improved gravity equation

$w_{ij}= K \frac{{M_i^\alpha}{M_j^\alpha}}{D_{ij}^\gamma}$

. The model maximizes the total expected traffic of the network. By changing the two parameters

α

and

γ

which controls the fitness and the geographical constraints, the model can vary its topology from the star-like network to the decentralized road-like network. The simulation for the Chinese city airline network reproduced many properties of the real network. In the end of this paper the relationship of the expected traffic and the real traffic is discussed.

Jiang-Hai Qian, Ding-Ding Han

A Bipartite Graph Based Model of Protein Domain Networks

Proteins are essential molecules of life in the cell and are involved in multiple and highly specialized tasks encoded in the amino acid sequence. In particular, protein function is closely related to fundamental units of protein structure called

domains

. Here, we investigate the distribution of kinds of domains in human cells. Our findings show that while the number of domain types shared by

k

proteins follows a scale-free distribution, the number of proteins composed of

k

types of domains decays as an exponential distribution. In contrast, previous data analyses and mathematical modeling reported a scale-free distribution for the protein domain distribution because the relation between kinds of domains and the number of domains in a protein was not considered. Based on this finding, we have developed an evolutionary model based on (1) growth process and (2) copy mechanism that explains the emergence of this mixing of exponential and scale-free distributions.

J. C. Nacher, T. Ochiai, M. Hayashida, T. Akutsu

The Results on the Stability of Glycolytic Metabolic Networks in Different Cells

Evolutionary forces will affect the structure of metabolic networks and their dynamic behaviors. To examine this hypothesis, in this work we investigate the relationship between the complexity of the metabolic glycolytic networks and the stability of the networks in different cells. By deriving the stoichiometrix from the FBA methods, we develop the models for Sce, Dmgr, Dsmi and Pic in fungi. Based on these models, we analyze the stability of the networks. The results show that the metabolic networks are more complicated with more stable ones.

Qinghua Zhou, Gang Peng, Li Jin, Momiao Xiong

The Probability Distribution of Inter-car Spacings

In this paper, the celluar automation model with Fukui-Ishibashi-type acceleration rule is used to study the inter-car spacing distribution for traffic flow. The method used in complex network analysis is applied to study the spacings distribution. By theoretical analysis, we obtain the result that the distribution of inter-car spacings follows power law when vehicle density is low and spacing is not large, while, when the vehicle density is high or the spacing is large, the distribution can be described by exponential distribution. Moreover, the numerical simulations support the theoretical result.

Jin Guo Xian, Dong Han

The Origin of Evolution in Physical Systems

A tentative outline for a model for the evolution of physical systems is presented. The universal classes of dynamical behaviors found in Cellular Automata experiments provide the basis for introducing the variation-stabiliza-tion principle as a synthetic interpretation of these phenomena. It is suggested that biological evolution takes its root in the evolution of physical systems as a particular case of the variation-stabilization principle that occurs at the transi-tion phase between ordered and chaotic regimes.

Jean-Claude Heudin

The Nonlinear Mechanism of Phase Transition in Computer Networks

In this paper, the nonlinear mechanism of phase transition in computer networks is analyzed, and a distributed proxy approach is introduced to improve network performance based on the two-dimensional coupling model. Theoretical analysis figures out that the nonlinear mechanism of router is the essential reason of network performance phase transition. Simulation results reveal that the extreme clustering characteristic of web access behavior gives arise to left-shift of phase transition critical compared with regular networks; after distributed proxy approach is employed, right-shift of the phase transition critical illustrates performance improvement. Finally, several important issues are mentioned.

Li Yi-Peng, Huang Yi-Hua, Wang Lei, Ren Yong

The Evolution of ICT Markets: An Agent-Based Model on Complex Networks

Information and communication technology (ICT) products exhibit positive network effects.The dynamic process of ICT markets evolution has two intrinsic characteristics: (1) customers are influenced by each others’ purchasing decision; (2) customers are intelligent agents with bounded rationality.Guided by complex systems theory, we construct an agent-based model and simulate on complex networks to examine how the evolution can arise from the interaction of customers, which occur when they make expectations about the future installed base of a product by the fraction of neighbors who are using the same product in his personal network.We demonstrate that network effects play an important role in the evolution of markets share, which make even an inferior product can dominate the whole market.We also find that the intensity of customers’ communication can influence whether the best initial strategy for firms is to improve product quality or expand their installed base.

Liangjie Zhao, Bangtao Wu, Zhong Chen, Li Li

The Effects of Link and Node Capacity on Traffic Dynamics in Weighted Scale-Free Networks

The effect of link and node capacity on traffic dynamics are investigated in weighted scale-free networks by adopting a traffic routing model with local node strength information:

$P_{l \rightarrow i} = \frac{S_{i}^{\alpha}}{\sum_j S_{j}^{\alpha}}$

. The link bandwidth is controlled by:

B

ij

 =  max (

βw

ij

,1), and the capacity of nodes is controlled by: max (

γs

i

, 1). The phase transition from free flow to congestion is reproduced. The optimal routing strategy is sought out. When

β

increases from zero, the optimal strategy changes from preferring low-strength nodes to high-strength nodes. When

β

 ≈ 1.0, there will be two optimal routing strategies. When

β

is low, the system’s behavior is controlled by link bandwidth, while it is controlled by node capacity when

β

is high. Our work may be useful for the design of modern traffic systems and communication networks.

M. B. Hu, R. Jiang, Y. H. Wu, Q. S. Wu

The Effect of Lane-Changing Time on the Dynamics of Traffic Flow

In this paper, the lane-changing time is considered in the cellular automata models for traffic flow. The lower the velocity of a vehicle, the longer the lane-changing time. The simulations are carried out in the two-lane system and the on-ramp system. When the lane-changing time is taken into account, the maximum flux per lane is reduced in the two-lane system compared with the original two-lane model, and it is even lower than that of single-lane road when a lane-changing takes longer time; the capacity drop can be reproduced in the on-ramp system.

Xin-Gang Li, Bin Jia, Rui Jiang

The Difference between Single-Valued and Multi-Valued Cases in the Compact Representation of CPD in Bayesian Networks

This paper addresses an important issue about the compact representation of the conditional probability distribution (CPD) applied in the well known Bayesian Networks in uncertain causality representation and probabilistic inference. That is, there is an essential difference between the single-valued cases and the multi-valued cases, while this difference does not exist when the CPD is represented in the conditional probability table (CPT). In other words, the present compact representation and inference methods applicable in the single-valued cases may not be applicable in the multi-valued cases as people usually think. A detailed example is provided to illustrate this problem. The solution is provided in the references by the author.

Qin Zhang

The Control Based on Internal Average Kinetic Energy in Complex Environment for Multi-robot System

In this paper, reference trajectory is designed according to minimum energy consumed for multi-robot system, which nonlinear programming and cubic spline interpolation are adopted. The control strategy is composed of two levels, which lower-level is simple PD control and the upper-level is based on the internal average kinetic energy for multi-robot system in the complex environment with velocity damping. Simulation tests verify the effectiveness of this control strategy.

Mao Yang, Yantao Tian, Xianghua Yin

The Contrast of Parametric and Nonparametric Volatility Measurement Based on Chinese Stock Market

Most procedures for modeling and forecasting financial asset return volatilities rely on restrictive and complicated parametric GARCH or stochastic volatility models. The method of realized volatility constructed from high-frequency intraday returns is an alternative choice for volatility measurement. In this paper we make an empirical analysis on Chinese stock index data by using the method of nonparametric realized volatility. We find that the realized volatility can describe the Chinese stock index volatility very well. The original Chinese stock index return series show obvious leptokurtic, fat-tailed relative to the Gaussian distribution.We show that the return series standardized instead by the realized volatility are very nearly Gaussian distribution, and we find that the four minutes is a better choice as the best time interval to describe the volatility of Chinese stock market. We also make a contrast with the popular method of GARCH model, but the return series standardized instead by GARCH model don’t accord with Gaussian distribution. The result shows that the realized volatility can describe the dynamic behaviors of Chinese stock market well. In a way, it indicates that the Chinese stock market is effective.

Xinwu Zhang, Yan Wang, Handong Li

The System Dynamics Research on the Private Cars’ Amount in Beijing

The thesis analyzes the development problem of private cars’ amount in Beijing from the perspective of system dynamics. With the flow chart illustrating the relationships of relevant elements, the SD model is established by VENSIM to simulate the growth trend of private autos’ amount in the future on the background of “Public Transportation First” policy based on the original data in Beijing. Then the article discusses the forecasting impacts of “Single-and-double license plate number limit” on the number of city vehicles and private cars under the assumption that this policy implemented for long after the 2008 Olympic Games. Finally, some recommendations are put forward for proper control over this problem.

Jie Fan, Guang-le Yan

The Topological Characteristics and Community Structure in Consumer-Service Bipartite Graph

We apply network analysis to study bipartite consumer- service graph that represents service transaction to understand consumer demand. Based on real-world computer log files of a library, we found that consumer graph projected from bipartite graph deviates significantly from theoretical predictions based on random bipartite graph. We observed smaller-than-expected average degree, larger-than-expected average path length and stronger-than-expected tendency to cluster. These findings motivated to explore the community structure of the network. As a result, the weighted consumer network showed significant community structure than the unweighted network. Communities picked out by the algorithm revealed that individuals in the same community were due to their common specialties or the overlapping structure of knowledge between their specialties.

Lin Li, Bao-Yan Gu, Li Chen

Time Dependent Virus Replication in Cell Cultures

We present in this report a stochastic model for the virus replication of influenza A in a cell culture. We consider not only the infection process of individual cells but also the number of intracellular components expressed in virus equivalent. Given that this expression is non constant in time we suggest a variable threshold, related to a viral resistance in the cell population, that could explain the time variation in the viral expression in the cell seen in experiments.

Juan G. Díaz Ochoa, Andreas Voigt, Heiko Briesen, Kai Sundmacher

You Never Walk Alone: Recommending Academic Events Based on Social Network Analysis

Combining Social Network Analysis and recommender systems is a challenging research field. In scientific communities, recommender systems have been applied to provide useful tools for papers, books as well as expert finding. However, academic events (conferences, workshops, international symposiums etc.) are an important driven forces to move forwards cooperation among research communities. We realize a SNA based approach for academic events recommendation problem. Scientific communities analysis and visualization are performed to provide an insight into the communities of event series. A prototype is implemented based on the data from DBLP and EventSeer.net, and the result is observed in order to prove the approach.

Ralf Klamma, Pham Manh Cuong, Yiwei Cao

Visualization of Complex Biological Systems: An Immune Response Model Using OpenGL

In this paper we present an update on our novel visualization technologies based on cellular immune interaction from both large-scale spatial and temporal perspectives. We do so with a primary motive: to present a visually and behaviourally realistic environment to the community of experimental biologists and physicians such that their knowledge and expertise may be more readily integrated into the model creation and calibration process. Visualization aids understanding as we rely on visual perception to make crucial decisions. For example, with our initial model, we can visualize the dynamics of an idealized lymphatic compartment, with antigen presenting cells (APC) and cytotoxic T lymphocyte (CTL) cells. The visualization technology presented here offers the researcher the ability to start, pause, zoom-in, zoom-out and navigate in 3-dimensions through an

idealised

lymphatic compartment.

John Burns, Heather J. Ruskin, Dimitri Perrin, John Walsh

Using the Weighted Rich-Club Coefficient to Explore Traffic Organization in Mobility Networks

The aim of a transportation system is to enable the movement of goods or persons between any two locations with the highest possible efficiency. This simple principle inspires highly complex structures in a number of real-world mobility networks of different kind that often exhibit a hierarchical organization. In this paper, we rely on a framework that has been recently introduced for the study of the management and distribution of resources in different real-world systems. This framework offers a new method for exploring the tendency of the top elements to form clubs with exclusive control over the system’s resources. Such tendency is known as the weighted rich-club effect. We apply the method to three cases of mobility networks at different scales of resolution: the US air transportation network, the US counties daily commuting, and the Italian municipalities commuting datasets. In all cases, a strong weighted rich-club effect is found. We also show that a very simple model can account for part of the intrinsic features of mobility networks, while deviations found between the theoretical predictions and the empirical observations point to the presence of higher levels of organization.

José J. Ramasco, Vittoria Colizza, Pietro Panzarasa

Tracking the Evolution in Social Network: Methods and Results

Contrary to previous static knowledge, our dynamic view in social network is so limited. Recent uncovering those hidden dynamic patterns has posed a series of challenging problems in network evolution. To make effective exploration, we present a fundamentally novel framework for uncovering the intricate properties of evolutionary networks. Different from static snapshots methods, we firstly trace the timelines of networks, which could explicitly characterize the network to several evolving segments. Then based on extracted smooth segments from the timeline, a graph approximation algorithm is devised to capture the frequent characteristics of the network and reduce the noise of interactions. Moreover, by employing the relationship between multi-attributes, an innovative community detection algorithm is proposed for detailed analysis on the approximate graphs. Besides the algorithms, to track these dynamic communities, we also introduce a community correlation and evaluation criterion. Finally, applying this framework to several synthetic and real-world datasets, we demonstrate the critical relationship between event and social evolution, and find that close-knit relationship with well-distributed tie strengths among members of large communities will contribute to a longer life span.

Shengqi Yang, Bin Wu, Bai Wang

Towards Network Complexity

In this paper, we briefly present a classification scheme of information-based network complexity measures. We will see that existing as well as novel measures can be divided into four major categories: (i) partition-based measures, (ii) non partition-based measures, (iii) non-parametric local measures and (iv) parametric local measures. In particular, it turns out that (ii)-(iv) can be obtained in polynomial time complexity because we use simple graph invariants, e.g., metrical properties of graphs. Finally, we present a generalization of existing local graph complexity measures to obtain parametric complexity measures.

Matthias Dehmer, Frank Emmert-Streib

Towards a Partitioning of the Input Space of Boolean Networks: Variable Selection Using Bagging

In this paper we present an algorithm that allows to select the input variables of Boolean networks from incomplete data. More precisely, sets of input variables, instead of single variables, are evaluated using mutual information to find the combination that maximizes the mutual information of input and output variables. To account for the incompleteness of the data bootstrap aggregation is used to find a stable solution that is numerically demonstrated to be superior in many cases to the solution found by using the complete data set all at once.

Frank Emmert-Streib, Matthias Dehmer

Toward Automatic Discovery of Malware Signature for Anti-Virus Cloud Computing

Security vendors are facing a serious problem of defeating the complexity of malwares. With the popularity and the variety of zero-day malware over the Internet, generating their signatures for detecting via anti-virus (AV) scan engines becomes an important reactive security function. However, AV security products consume much of the PC memory and resources due to their large signature files. AV cloud computing becomes a popular solution for this problem. In this paper, a novel Automatic Malware Signature Discovery System for AV cloud (AMSDS) is proposed to generate malware signatures from both static and dynamic aspects. Our experiments on millions-scale samples suggest that AMSDS outperforms most state-of-the-art automatic signature generation techniques of both industry and academia.

Wei Yan, Erik Wu

Topological Structure and Interest Spectrum of the Group Interest Network

In this paper, the behavior characteristics that the specifical campus group users accessing world wide web has been studied, the dynamic group interest network has been constructed, which was a para-bipartite graph and the topological structure had been discussed. Although the users’ visiting time is random and the web pages they visited are different but the interests of a majority of the campus group are accordant. The results indicate that the incoming degree distribution of the group interest network follows power law. And the group interest spectrum was basically steady. The visiting behavior of the campus group had their special disciplinarian.

Ning Zhang

Topological Analysis and Measurements of an Online Chinese Student Social Network

Online social network attracts more researchers now. In this paper, we topologically analyze an online Chinese student social network–Xiaonei.com. We use Python language to crawl two datasets of Xiaonei in January and February, 2008. The degree distribution and small world phenomena are testified. We also use a social network analysis tool to analyze these two datasets from the viewpoint of social network structure. Seventeen measurements such as Fragmentation, Component Count, Strong/Weak are summarized to identify the exogenous attributes of Xiaonei.com. Additionally, two latent applications of online social network service are proposed in the discussion section.

Duoyong Sun, Jiang Wu, Shenghua Zheng, Bin Hu, Kathleen M. Carley

Time, Incompleteness and Singularity in Quantum Cosmology

In this paper we extend our 2007 paper, “Comparative Quantum Cosmology: Causality, Singularity, and Boundary Conditions”, http://arxiv.org/ftp/arxiv/papers/0710/0710.5046.pdf, to include consideration of universal expansion, various implications of extendibility and incompleteness in spacetime metrics and, absent the treatment of Feynman diagrams, the use of Penning trap dynamics to describe the Hamiltonians of space-times with no characteristic upper or lower bound.

Philip V. Fellman, Jonathan Vos Post, Christine Carmichael, Alexandru Manus, Dawna Lee Attig

The Complex Economic System of Supply Chain Financing

Supply Chain Financing (SCF) refers to a series of innovative and complicated financial services based on supply chain. The SCF set-up is a complex system, where the supply chain management and Small and Medium Enterprises (SMEs) financing services interpenetrate systematically. This paper establishes the organization structure of SCF System, and presents two financing models respectively, with or without the participation of the third-party logistic provider (3PL). Using Information Economics and Game Theory, the interrelationship among diverse economic sectors is analyzed, and the economic mechanism of development and existent for SCF system is demonstrated. New thoughts and approaches to solve SMEs financing problem are given.

Lili Zhang, Guangle Yan

The Bipartite Network Study of the Library Book Lending System

Through collecting the library lending information of the University of Shanghai for Science and Technology during one year, we build the database between the books and readers, and then construct a bipartite network to describe the relationships. We respectively establish the corresponding un-weighted and weighted bipartite network through the borrowing relationship and the reading days, thereout obtain the statistical properties via the theory and methods of complex network. We find all the properties follow exponential distribution and there is a positive correlation between the relevant properties in un-weighted and weighted networks. The un-weighted properties can describe the cooperation situation and configuration, but the properties with node weight may describe the competition results. Besides, we discuss the practical significance for the double relationship and the statistical properties. Further more, we propose a library personal recommendation system for developing the library humanity design resumptively.

Nan-nan Li, Ning Zhang

Temperature-Induced Domain Shrinking in Ising Ferromagnets Frustrated by a Long-Range Interaction

We investigate a spin model in which a ferromagnetic short-range interaction competes with a long-range antiferromagnetic interaction decaying spatially as

$\frac{1}{r^{d+\sigma}}$

,

d

being the dimensionality of the lattice. For

σ

smaller than a certain threshold

$\hat{\sigma}$

(with

$\hat{\sigma}>1$

), the long-range interaction is able to prevent global phase separation, the uniformly magnetized state favored by the exchange interaction for spin systems. The ground state then consists of a mono-dimensional modulation of the order parameter resulting in a superlattice of domains with positive and negative magnetization. We find that the period of modulation shrinks with increasing temperature

T

and suggest that this is a universal property of the considered model. For

d

 = 2 and

σ

= 1 (dipolar interaction) Mean-Field (MF) calculations find a striking agreement with experiments performed on atomically-thin Fe/Cu(001) films. Monte Carlo (MC) results for

d

 = 1 also support the generality of our arguments beyond the MF approach.

Alessandro Vindigni, Oliver Portmann, Niculin Saratz, Fabio Cinti, Paolo Politi, Danilo Pescia

Slowdown in the Annihilation of Two Species Diffusion-Limited Reaction on Fractal Scale-Free Networks

In the diffusion-limited reaction process

A

 + 

B

→ ∅ on random scale-free networks, particle density decays as

ρ

(

t

) ~

t

 − 

α

when

ρ

A

(0) = 

ρ

B

(0), where

α

> 1 for the degree exponent 2 < 

γ

< 3 and

α

= 1 for

γ

 ≤ 3. We investigate the reaction on fractal scale-free networks numerically, finding

ρ

(

t

) decays slowly with the exponent

α

 ≈ 

d

s

/ 4 < 1, where

d

s

is the spectral dimension of the network.

Chang-Keun Yun, Byungnam Kahng, Doochul Kim

SIRS Dynamics on Random Networks: Simulations and Analytical Models

The standard pair approximation equations (PA) for the Susceptible-Infective-Recovered-Susceptible (SIRS) model of infection spread on a network of homogeneous degree

k

predict a thin phase of sustained oscillations for parameter values that correspond to diseases that confer long lasting immunity. Here we present a study of the dependence of this oscillatory phase on the parameter

k

and of its relevance to understand the behaviour of simulations on networks. For

k

 = 4, we compare the phase diagram of the PA model with the results of simulations on regular random graphs (RRG) of the same degree. We show that for parameter values in the oscillatory phase, and even for large system sizes, the simulations either die out or exhibit damped oscillations, depending on the initial conditions. This failure of the standard PA model to capture the qualitative behaviour of the simulations on large RRGs is currently being investigated.

Ganna Rozhnova, Ana Nunes

Self-organized Collaboration Network Model Based on Module Emerging

Recently, the studies of the complex network have gone deep into many scientific fields, such as computer science, physics, mathematics, sociology, etc. These researches enrich the realization for complex network, and increase understands for the new characteristic of complex network. Based on the evolvement characteristic of the author collaboration in the scientific thesis, a self-organized network model of the scientific cooperation network is presented by module emerging. By applying the theoretical analysis, it is shown that this network model is a scale-free network, and the strength degree distribution and the module degree distribution of the network nodes have the same power law. In order to make sure the validity of the theoretical analysis for the network model, we create the computer simulation and demonstration collaboration network. By analyzing the data of the network, the results of the demonstration network and the computer simulation are consistent with that of the theoretical analysis of the model.

Hongyong Yang, Lan Lu, Qiming Liu

Self-organized Balanced Resources in Random Networks with Transportation Bandwidths

We apply statistical physics to study the task of resource allocation in random networks with limited bandwidths for the transportation of resources along the links. We derive algorithms which searches the optimal solution without the need of a global optimizer. For networks with uniformly high connectivity, the resource shortage of a node becomes a well-defined function of its capacity. An efficient profile of the allocated resources is found, with clusters of node interconnected by an extensive fraction of unsaturated links, enabling the resource shortages among the nodes to remain balanced. The capacity-shortage relation exhibits features similar to the Maxwell’s construction. For scale-free networks, such an efficient profile is observed even for nodes of low connectivity.

Chi Ho Yeung, K. Y. Michael Wong

Selection of Imitation Strategies in Populations When to Learn or When to Replicate?

A question in the modeling of populations of imitators is if simple imitation or imitation based on learning rules can improve the fitness of the individuals. In this investigation this problem is analyzed for two kinds of imitators involved in a cooperative dilemma: One kind of imitators has a replicator heuristics, i.e. individuals which decide its new action based on actions of their neighbors, whereas a second type has a learning heuristics, i.e. individuals which use a learning rule (for short learner) in order to determine their new action. The probability that a population of learners penetrates in a population of replicators depends on a training error parameter assigned to the replicators. I show that this penetration is similar to a site percolation process which is robust to changes in the individual learning rule.

Juan G. Díaz Ochoa

Sediment Transport Dynamics in River Networks: A Model for Higher-Water Seasons

A dynamical model is proposed to study sediment transport in river networks in higher-water seasons. The model emphasizes the difference between the sediment-carrying capability of the stream in higher-water seasons and that in lower-water seasons. The dynamics of sediment transport shows some complexities such as the complex dependence of the sediment-carrying capability on sediment concentration, the response of the channel(via erosion or sedimentation) to the changes of discharge.

Jie Huo, Xu-Ming Wang, Rui Hao, Jin-Feng Zhang

Scaling Relations in Absorbing Phase Transitions with a Conserved Field in One Dimension

Validity of two scaling relations

β

 = 

ν

 ∥ 

θ

and

z

 = 

ν

 ∥ 

/

ν

 ⊥ 

widely known in absorbing phase transitions is studied for the conserved lattice gas (CLG) model and the conserved threshold transfer process CTTP) both in one dimension. For the CLG model, it is found that both relations hold when the critical exponents calculated from the all-sample average density of active particles are considered. For the CTTP model, various exponents are calculated via Monte Carlo simulations and they are confirmed by the off-critical scaling and the finite-size scaling analyses. The exponents estimated from the all-sample averages again satisfy both relations. These observations are in strict disagreement with earlier observations in two dimensions [Phys. Rev. Lett.

85

, 1803 (2000); Phys Rev. E

68

, 056102 (2003)] but support the more recent observation for the CLG model [Phys. Rev. E

78

, 040103(R) (2008)].

Sang-Gui Lee, Sang Bub Lee

Scaling Law between Urban Electrical Consumption and Population in China

The relation between the household electrical consumption

Y

and population

N

for Chinese cities in 2006 has been investigated with the power law scaling form

$Y = A_0 N^{\beta}$

. It is found that the Chinese cities should be divided into three categories characterized by different scaling exponent

β

. The first category, which includes the biggest and coastal cities of China, has the scaling exponent

β

> 1. The second category, which includes mostly the cities in central China, has the scaling exponent

β

 ≈ 1. The third category, which consists of the cities in northwestern China, has the scaling exponent

β

< 1 . Using a urban growth equation, different ways of city population evolution can be obtained for different

β

. For

β

< 1 , population evolutes always to a fixed point population

N

f

from below or above depending on the initial population. For

β

> 1, there is also a fixed point population

N

f

. If the initial population

N

(0) > 

N

f

, the population increases very fast with time and diverges within a finite time. If the initial population

N

(0) < 

N

f

, the population decreases with time and collapse finally. The pattern of population evolution in a city is determined by its scaling exponent and initial population.

Xiaowu Zhu, Aimin Xiong, Liangsheng Li, Maoxin Liu, X. S. Chen

Scaling in Modulated Systems

How can we understand a system that is too complicated to be simulated and in which some important quantities cannot be determined from observation? This is a question we were confronted with when we analyzed experimentally observed magnetic patterns in ultrathin ferromagnetic films. We have now, for this specific case, found a method that gives a good qualitative understanding of a surprising reentrance of order observed experimentally. This method is based on scaling arguments and may prove useful in the study of other complex systems.

Oliver Portmann, Alessandro Vindigni, Danilo Pescia

Scaling Behavior of Chinese City Size Distribution

We have investigated the population distribution of Chinese cities from 1997 to 2006. The rank-size distributions of Chinese cities deviate from the Pareto distribution. For city size distribution of each year we can find a population threshold

P

c

that characterizes the boundary of the deviation. The cities with population more than

P

c

follow the Pareto distribution, while the smaller cities deviate from the Pareto distribution. Using

P

c

for every year, the rank-size distribution from 1997 to 2006 can be written into a scaling form

$R(P,T)= C(T)P^{-\alpha (T)}f(P/P_c (T))$

, where the Pareto exponent

α

(

T

) is not equal to the value of Zipf’s law and evolutes with time. According this scaling form, the data of the city size distributions of Chinese cities from 1997 to 2006 can collapses to a single curve, which is the scaling function of the city size distribution.

Xiaowu Zhu, Aimin Xiong, Liangsheng Li, Maoxin Liu, Xiaosong Chen

Social Network as Double-Edged Sword to Exchange: Frictions and the Emerging of Intellectual Intermediary Service

The value of complex social network and the optimization of it are determined by the structure and nodes’ characteristics. Direct friction and indirect friction are defined to describe the possible exchange difficulty each node meets with its neighbors in exchange network. Exogenous intermediary and endogenous intermediary can decrease these frictions by adding links. Agent-based Simulating results show that both frictions and the optimization of them are influenced by demander and supplier rate, the exchange network structure as well as the environment constrains and exogenous intermediation acts better than endogenous intermediation in decreasing both frictions. While assists exchange, the results of this paper also implies social network as origin of impefect market.

Li Li, Bangtao Wu, Zhong Chen, Liangjie Zhao

Spam Source Clustering by Constructing Spammer Network with Correlation Measure

Spam filtering is one of the most challenging problems in electric message systems. In general, recent studies on specifying real spam source are based on content filtering because spammers usually falsify their origin. We propose a method to specify spam source based on structural analysis with complex network. We assume that each spam sources either has the same victim list or uses the same spam-hosting program. We treat spam source - target relationship as a bipartite network and construct weighted spam source network by network projection using correlation measure. We find that community clustering methods are inappropriate with spammer network. We group spammers with gradient-based grouping, which uses correlations between nodes as gradient between nodes. We convert them into local minima, which helps to cluster spammers into a few spam source groups. We investigate the weblog spam data with the proposed method and validate it. The method that we propose can be applied to diverse categorization problems, such as multiple text categorization and network subunit clustering.

Jeongkyu Shin, Seunghwan Kim

Spiral Waves Emergence in a Cyclic Predator-Prey Model

Based on a cyclic predator-prey model of three species, spiral waves on global level of the system are obtained. It is found that the predation intensity greatly affects on the behaviors of spiral waves. The wavelength of spiral waves alter with the mobility in the form of

λ

~

D

θ

. Values of

θ

are determined by predation rates between species. It indicates the behaviors of spiral waves varying with mobility are universal at the same predation rate which reveals competition of resources among species.

Luo-Luo Jiang, Wen-Xu Wang, Xin Huang, Bing-Hong Wang

Synchronization Stability of Coupled Near-Identical Oscillator Network

To study the effect of parameter mismatch on the stability in a general fashion, we derive variational equations to analyze the stability of synchronization for coupled near-identical oscillators. We define master stability equations and associated master stability functions, which are independent of the network structure. In particular, we present several examples of coupled near-identical Lorenz systems configured in small networks (a ring graph and sequence networks) with a fixed parameter mismatch and a large Barabasi-Albert scale-free network with random parameter mismatch. We find that several different network architectures permit similar results despite various mismatch patterns.

abstract

environment.

Jie Sun, Erik M. Bollt, Takashi Nishikawa

Synchronization of Complex Networks with Time-Varying Coupling Delay via Impulsive Control

Impulsive control and exponential synchronization analysis of a class of complex networks with time-varying coupling delay is investigated in this paper. Our aim is to enhance the synchronizability of the complex networks by applying impulsive control. By introducing a comparison system and estimating the corresponding Cauchy matrix sufficient conditions on global exponential synchronization are derived. An impulsive controller is explicitly designed not only to achieve synchronized dynamics for the complex networks, but simultaneously to ensure the states of synchronous error converging with a given decay rate. A numerical example is presented to illustrate the theoretical results and proposed controller design procedure.

Yang Dai, Yunze Cai, Xiaoming Xu

Synchronization in Complex Networks with Different Sort of Communities

In this paper, inspired by the idea that many real networks are composed by sorts of communities, we investigate the synchronization property of oscillators on such community networks. We identify the communities by two ways, one is by the structure of individual community and the other by the intrinsic frequencies probability density

g

(

ω

) of Kuramoto oscillators on different communities. For the two sorts of community networks, when the community structure is strong, only the oscillators on the same community synchronize. With the weakening of the community strength, an interesting phenomenon appears: although the global synchronization is not achieved, oscillators on the same sort of communities will synchronize independently. Global synchronization will appear with the further weakening of community structure.

Ming Zhao, Tao Zhou, Hui-Jie Yang, Gang Yan, Bing-Hong Wang

Symmetry Breaking in the Evolution of World Economic Structure

Over the centuries, world economic system and the corresponding economic structure have been in a state of continuous evolution. In this paper, through the empirical analysis on the evolution history of world economic structure, we show that the underlying driving force for the evolution of world economic structure is Technology Innovation. Specifically, we find that symmetry breakings not only emerge in the whole economic structure, but also take place in the local economic relation and economic status of inner countries along the long evolution history of world economic structure. We also elaborate the detailed mechanism of symmetry breaking of world economic structure. That is, in the evolution of world economic structure, all those countries participating in world economic open market are affected to varying degrees by symmetry breaking that is caused by technology innovation, which eventually determines current world economic structure with competitive countries evolving into economic centers and countries completely marginalized evolving into ‘singularities’ of world economic network.

Hui Wang, Guangle Yan

Studies on Interpretive Structural Model for Forest Ecosystem Management Decision-Making

Characterized by their openness, complexity and large scale, forest ecosystems interweave themselves with social system, economic system and other natural ecosystems, thus complicating both their researches and management decision-making. According to the theories of sustainable development, hierarchy-competence levels, cybernetics and feedback, 25 factors have been chosen from human society, economy and nature that affect forest ecosystem management so that they are systematically analyzed via developing an interpretive structural model (ISM) to reveal their relationships and positions in the forest ecosystem management. The ISM consists of 7 layers with the 3 objectives for ecosystem management being the top layer (the seventh layer). The ratio between agricultural production value and industrial production value as the bases of management decision-making in forest ecosystems becomes the first layer at the bottom because it has great impacts on the values of society and the development trends of forestry, while the factors of climatic environments, intensive management extent, management measures, input-output ratio as well as landscape and productivity are arranged from the second to sixth layers respectively.

Suqing Liu, Xiumei Gao, Qunying Zen, Yuanman Zhou, Yuequn Huang, Weidong Han, Linfeng Li, Jiping Li, Yingshan Pu

Structure of Mutualistic Complex Networks

We consider the structures of six plant-pollinator mutualistic networks. The plants and pollinators are linked by the plant-pollinating relation. We assigned the visiting frequency of pollinators to a plant as a weight of each link. We calculated the cumulative distribution functions of the degree and strength for the networks. We observed a power-law, linear, and stretched exponential dependence of the cumulative distribution function. We also calculated the disparity and the strength of the nodes

s

(

k

) with degree

k

. We observed that the plant-pollinator networks exhibit an disassortative behaviors and nonlinear dependence of the strength on the nodes. In mutualistic networks links with large weight are connected to the neighbors with small degrees.

Jun Kyung Hwang, Seong Eun Maeng, Moon Yong Cha, Jae Woo Lee

Strong Dependence of Infection Profiles on Grouping Dynamics during Epidemiological Spreading

The spreading of an epidemic depends on the connectivity of the underlying host population. Because of the inherent difficulties in addressing such a problem, research to date on epidemics in networks has focused either on static networks, or networks with relatively few rewirings per timestep. Here we employ a simple, yet highly non-trivial, model of dynamical grouping to investigate the extent to which the underlying dynamics of tightly-knit communities can affect the resulting infection profile. Individual realizations of the spreading tend to be dominated by large peaks corresponding to infection resurgence, and a generally slow decay of the outbreak. In addition to our simulation results, we provide an analytical analysis of the run-averaged behaviour in the regime of fast grouping dynamics. We show that the true run-averaged infection profile can be closely mimicked by employing a suitably weighted static network, thereby dramatically simplifying the level of difficulty.

Zhenyuan Zhao, Guannan Zhao, Chen Xu, Pak Ming Hui, Neil F. Johnson

Statistical Properties of Cell Topology and Geometry in a Tissue-Growth Model

Statistical properties of cell topologies in two-dimensional tissues have recently been suggested to be a consequence of cell divisions. Different rules for the positioning of new walls in plants have been proposed, where e.g. Errara’s rule state that new walls are added with the shortest possible path dividing the mother cell’s volume into two equal parts. Here, we show that for an isotropically growing tissue Errara’s rule results in the correct distributions of number of cell neighbors as well as cellular geometries, in contrast to a random division rule. Further we show that wall mechanics constrain the isotropic growth such that the resulting cell shape distributions more closely agree with experimental data extracted from the shoot apex of

Arabidopsis thaliana

.

Patrik Sahlin, Olivier Hamant, Henrik Jönsson

Stability of Non-diagonalizable Networks: Eigenvalue Analysis

The stability of non-diagonalizable networks of dynamical systems are investigated in detail based on eigenvalue analysis. Pinning control is suggested to stabilize the synchronization state of the whole coupled network. The complicated coupled problem is reduced to two independent problems: clarifying the stable region of the modified system and specifying the eigenvalue distribution of the coupling and control matrix. The dependence of the stability on both pinning density and pinning strength is studied.

Linying Xiang, Zengqiang Chen, Jonathan J. H. Zhu

Scale-Free Networks with Different Types of Nodes

In many natural and social networks, nodes may play different roles or have different functions. In this paper, we propose a simple model with different types of nodes and deterministic selective linking rule. We investigate the structural properties by theoretical predictions. It is found that the given model exhibits a power-law distribution. In addition, we make the model become the weighted network by giving the links the weight and analyze the probability distribution of the node strength.

Juan Zhang, Wenfeng Wu

Global Synchronization of Generalized Complex Networks with Mixed Coupling Delays

In this paper we propose a generalized complex networks model, which concerns asymmetric network configuration including both neutral-type coupling delay and retarded-type one. The synchronization problem of this generalized complex networks is reformulated into the asymptotical stability problem of neutral delay functional differential equations. By introducing descriptor system transformation strategy, the less conservative sufficient condition of delay-independent and independent-of-delay global synchronization criteria are derived in terms of linear matrix inequalities. A numerical example is given to support the theoretical results.

Yang Dai, Yunze Cai, Xiaoming Xu

Community Division of Heterogeneous Networks

Many real world data can be represented as heterogeneous networks that are composed of more than one types of nodes, such as paper-author networks (two types) and user-resource-tag networks (three types) of social tagging systems. Discovering communities from such heterogeneous networks is important for finding similar nodes, which are useful for information recommendation and visualization. Although modularity is a famous criterion for evaluating division of given networks, it is not applicable to heterogeneous networks. This paper proposes new modularity for bipartite networks, as the first step for heterogeneous networks. Experimental results using artificial networks and real networks are shown.

Tsuyoshi Murata

Autonomous Co-operation and Control in Complex Adaptive Logistic Systems – Contributions and Limitations for the Innovation Capability of International Supply Networks

This paper aims to analyze the potential contributions of the organization principle autonomous co-operation and control to the innovation capabilities of logistics systems and their sub-systems like single organizations. Therefore, the concept of Complex Adaptive Logistics Systems (CALS) will be introduced and the essentiality of the heterogeneity of the elements within logistics systems for their innovation capabilities will be emphasized. One possible driver for homogeneity is the so-called dominant logic.

Michael Hülsmann, Philip Cordes

Asymptotic Behavior of Ruin Probability in Insurance Risk Model with Large Claims

For the renewal risk model with subexponential claim sizes, we established for the finite time ruin probability a lower asymptotic estimate as initial surplus increases, subject to the demand that it should hold uniformly over all time horizons in an infinite interval. In the case of Poisson model, we also obtained the upper asymptotic formula so that an equivalent formula was derived. These extended a recent work partly on the topic from the case of Pareto-type claim sizes to the case of subexponential claim sizes and, simplified the proof of lower bound in Leipus and Siaulys ([9]).

Tao Jiang

Approaching the Linguistic Complexity

We analyze the rank-frequency distributions of words in selected English and Polish texts. We compare scaling properties of these distributions in both languages. We also study a few small corpora of Polish literary texts and find that for a corpus consisting of texts written by different authors the basic scaling regime is broken more strongly than in the case of comparable corpus consisting of texts written by the same author. Similarly, for a corpus consisting of texts translated into Polish from other languages the scaling regime is broken more strongly than for a comparable corpus of native Polish texts. Moreover, based on the British National Corpus, we consider the rank-frequency distributions of the grammatically basic forms of words (lemmas) tagged with their proper part of speech. We find that these distributions do not scale if each part of speech is analyzed separately. The only part of speech that independently develops a trace of scaling is verbs.

Stanisław Drożdż, Jarosław Kwapień, Adam Orczyk

Application of the Kelly Criterion to Ornstein-Uhlenbeck Processes

In this paper, we study the Kelly criterion in the continuous time framework building on the work of E.O. Thorp and others. The existence of an optimal strategy is proven in a general setting and the corresponding optimal wealth process is found. A simple formula is provided for calculating the optimal portfolio for a set of price processes satisfying some simple conditions. Properties of the optimal investment strategy for assets governed by multiple Ornstein-Uhlenbeck processes are studied. The paper ends with a short discus-sion of the implications of these ideas for financial markets.

Yingdong Lv, Bernhard K. Meister

Application of SRM to Diverse Populations

In today modern industrial cities we see that many people having different cultures share the same settlement and form a typical social complex system. People come from other cities or even foreign countries Newcomers bring their cultural values such as clothing, meals, likes and dislikes. As a result of interacting with other people some cultural values change, some completely forgotten while others become popular and known by the majority of people. There should be a mechanism helping some cultural values being more popular and causing other people being assimilated by majorities. Different cultures’ interactions with each other and consequences of their interactions will be investigated by the principle rules of Simple Recommendation Model which is proposed by Bingol in 2006. The agents will be grouped according to their national origin and remember and forget the choices instead of agents. Also selections of interacted agents will be made according to people’s choices.

Sahin Delipinar, Haluk Bingol

Antisynchronization of Two Complex Dynamical Networks

A nonlinear type open-plus-closed-loop (OPCL) coupling is investi-gated for antisynchronization of two complex networks under unidirectional and bidirectional interactions where each node of the networks is considered as a continuous dynamical system. We present analytical results for antisynchroni-zation in identical networks. A numerical example is given for unidirectional coupling with each node represented by a spiking-bursting type Hindmarsh-Rose neuron model. Antisynchronization for mutual interaction is allowed only to inversion symmetric dynamical systems as chosen nodes.

Ranjib Banerjee, Ioan Grosu, Syamal K. Dana

Analysis and Modeling on the Government’s Co-agglomeration in Industrial Clustering

Industry clusters have been the focus of scholars and governments since the second half of the 20th century. During a cluster’s growing process, the government plays an important role. In order to show the growing law of the clusters and how government did for co-agglomeration, we proposed two kinds of models: Logistic model and BA model with parameter

α

to describe the single and mass clusters separately, and we choose the gross industrial output value of the 13 cities in Jiangsu province as numerical verification, showing that the government is part and parcel of the industrial clusters.

Ying-Chao Zhang, Chao Chen, Xin-Yi Huang, Xiao-Ling Ye, Yi-Lu Cai

Analysing Weighted Networks: An Approach via Maximum Flows

We present an approach for analysing weighted networks based on maximum flows between nodes and generalize to weighted networks ‘global’ measures that are well-established for binary networks, such as pathlengths, component size or betweenness centrality. This leads to a generalization of the algorithm of Girvan and Newman for community identification. The application of the weighted network measures to two real-world example networks, the international trade network and the passenger flow network between EU member countries, demonstrates that further insights about the systems’ architectures can be gained this way.

Markus Brede, Fabio Boschetti

An Emergence Principle for Complex Systems

From elementary system graph representation, systems are shown to belong to only three states: simple, complicated, and complex. First two have been studied over past centuries. Last one originates in existence of threshold above which components interaction overtakes outside interaction, leading to system self-organization which filters outer action, making it more robust with emergence of new behaviour not predictable from components study. The threshold value, expressed in terms of coupling system parameters, is verified to recovers limits found in a broad range of domains in Physics and Mathematics, giving explicit criterion for emergence in complex system. Application to man-made systems concentrates on the balance between relative system isolation when becoming complex and delegation of more “intelligence” in adequate frame between new augmented system state and supervising operator. Entering complexity state opens the possibility for the function to feedback onto the structure, ie to mimic technically the early invention of Nature.

Michel Cotsaftis

An Effective Local Routing Strategy on the Communication Network

In this paper, we propose an effective routing strategy on the basis of the so-called nearest neighbor search strategy by introducing a preferential cut-off exponent

K

. We assume that the handling capacity of one vertex is proportional to its degree when the degree is smaller than

K

, and is a constant

C

0

otherwise. It is found that by tuning the parameter

α

, the scale-free network capacity measured by the order parameter is considerably enhanced compared to the normal nearest-neighbor strategy. Traffic dynamics both near and far away from the critical generating rate

R

c

are discussed. Simulation results demonstrate that the optimal performance of the system corresponds to

α

= − 0.5. Due to the low cost of acquiring nearest-neighbor information and the strongly improved network capacity, our strategy may be useful and reasonable for the protocol designing of modern communication networks.

Yu-jian Li, Bing-hong Wang, Zheng-dong Xi, Chuan-yang Yin, Han-xin Yang, Duo Sun

Average Consensus in Delayed Networks of Dynamic Agents with Impulsive Effects

In this paper, the issues of average consensus in undirected delayed networks of dynamic agents with impulsive effects are investigated. The primary contribution of this paper is to propose the consensus schemes in undirected delayed networks of dynamic agents having impulsive effects as well as fixed, switching topology. Based on impulsive stability theory on delayed dynamical systems, we derive some simple sufficient conditions under which all the nodes in the network achieve average consensus globally exponentially. It is shown that average consensus in the networks is heavily dependent on impulsive effects of communication topology of the networks. Subsequently, two numerical examples illustrate and visualize the effectiveness and feasibility of our theoretical results.

Quanjun Wu, Lan Xiang, Jin Zhou

Basic Notions and Models in Systems Science

The development of the idea of seeing parts of the world as ‘related objects’ or the ‘systemic view’ and its relation to conventional science is briefly described. Concepts in the systemic view regarded as fundamental and their expression as linguistic and mathematical models which would turn this view into ‘systems science’, are introduced. Products are represented as sets and linguistic networks of ordered pairs. Semantic diagrams describe the dynamics of change. A case study to illustrate the basic notions and models is given.

Janos Korn

Bifurcation Phenomena of Opinion Dynamics in Complex Networks

In this paper, we study the opinion dynamics of Improved Deffuant model (IDM), where the convergence parameter

μ

is a function of the opposite’s degree

K

according to the celebrity effect, in small-world network (SWN) and scale-free network (SFN). Generically, the system undergoes a phase transition from the plurality state to the polarization state and to the consensus state as the confidence parameter

ε

increasing. Furthermore, the evolution of the steady opinion

s

*

as a function of

ε

, and the relation between the minority steady opinion

$s_{*}^{min}$

and the individual connectivity

k

also have been analyzed. Our present work shows the crucial role of the confidence parameter and the complex system topology in the opinion dynamics of IDM.

Long Guo, Xu Cai

Community Detection of Time-Varying Mobile Social Networks

In this paper, we present our ongoing work on developing a framework for detecting time-varying communities on human mobile networks. We define the term

community

in environments where the mobility patterns and clustering behaviors of individuals vary in time. This work provides a method to describe, analyze, and compare the clustering behaviors of collections of mobile entities, and how they evolve over time.

Shu-Yan Chan, Pan Hui, Kuang Xu

Collaborative Transportation Planning in Complex Adaptive Logistics Systems: A Complexity Science-Based Analysis of Decision-Making Problems of “Groupage Systems”

This paper aims to analyze decision-making problems in Groupage Systems from a complexity-science perspective. Therefore, the idea of Complex Adaptive Logistics Systems (CALS) and its inherent organization principle of autonomous co-operation and control will be presented. Furthermore, Groupage systems as a way to implement collaborative transportation planning will be introduced and, in combination with the idea of CALS, resulting decisionmaking problems for so-called ‘smart parts’ in logistics systems will be deduced.

Michael Hülsmann, Herbert Kopfer, Philip Cordes, Melanie Bloos

Classification Based on the Optimal K-Associated Network

In this paper, we propose a new graph-based classifier which uses a special network, referred to as optimal

K-associated network

, for modeling data. The

K

-associated network is capable of representing (dis)similarity relationships among data samples and data classes. Here, we describe the main properties of the

K

-associated network as well as the classification algorithm based on it. Experimental evaluation indicates that the model based on an optimal

K

-associated network captures topological structure of the training data leading to good results on the classification task particularly for noisy data.

Alneu A. Lopes, João R. Bertini, Robson Motta, Liang Zhao

Characterizing the Structural Complexity of Real-World Complex Networks

Although recent research has shown that the complexity of a network depends on its structural organization, which is linked to the functional constraints the network must satisfy, there is still no systematic study on how to distinguish topological structure and measure the corresponding structural complexity of complex networks. In this paper, we propose the first consistent framework for distinguishing and measuring the structural complexity of real-world complex networks. In terms of the smallest

d

of the

dK

model with high-order constraints necessary for fitting real networks, we can classify real-world networks into different structural complexity levels. We demonstrate the approach by measuring and classifying a variety of real-world networks, including biological and technological networks, small-world and non-small-world networks, and spatial and non-spatial networks.

Jun Wang, Gregory Provan

Backmatter

Additional information

Premium Partner

    Image Credits