Bonan Hou

National University of Defense Technology, Ch’ang-sha-shih, Hunan, China

Are you Bonan Hou?

Claim your profile

Publications (10)2.85 Total impact

  • Bonan Hou · Yiping Yao · Dongsheng Liao ·
    [Show abstract] [Hide abstract]
    ABSTRACT: Identifying the most influential nodes in complex networks provides a strong basis for understanding spreading dynamics and ensuring more efficient spread of information. Due to the heterogeneous degree distribution, we observe that current centrality measures are correlated in their results of nodes ranking. This paper introduces the concept of all-around nodes, which act like all-around players with good performance in combined metrics. Then, an all-around distance is presented for quantifying the influence of nodes. The experimental results of susceptible-infectious-recovered (SIR) dynamics suggest that the proposed all-around distance can act as a more accurate, stable indicator of influential nodes.
    Physica A: Statistical Mechanics and its Applications 08/2012; 391(15):4012–4017. DOI:10.1016/j.physa.2012.02.033 · 1.73 Impact Factor
  • Bing Wang · Bonan Hou · Fei Xing · Yiping Yao ·
    [Show abstract] [Hide abstract]
    ABSTRACT: The spatial stochastic simulation of biochemical systems requires significant calculation efforts. Parallel discrete-event simulation is a promising approach to accelerate the execution of simulation runs. However, achievable speedup depends on the parallelism inherent in the model. One of our goals is to explore this degree of parallelism in the Next Subvolume Method type simulations. Therefore we introduce the Abstract Next Subvolume Method, in which we decouple the model representation from the sequential simulation algorithms, and prove that state trajectories generated by its executions statistically accord with those generated by the Next Subvolume Method. The experimental performance analysis shows that optimistic synchronization algorithms, together with careful controls over the speculative execution, are necessary to achieve considerable speedup and scalability in parallel spatial stochastic simulation of chemical reactions. Our proposed method facilitates a flexible incorporation of different synchronization algorithms, and can be used to select the proper synchronization algorithm to achieve the efficient parallel simulation of chemical reactions.
    Computational biology and chemistry 06/2011; 35(3):193-8. DOI:10.1016/j.compbiolchem.2011.05.001 · 1.12 Impact Factor
  • Bonan Hou · Yiping Yao · Shaoliang Peng ·
    [Show abstract] [Hide abstract]
    ABSTRACT: The entity interaction graph is an important metaphor for understanding the simulation execution of complex systems on parallel computing environment. Current perfor- mance tuning techniques often explore interrelated factors affect- ing performance, but ignore systematic analysis on the structure and behavior of entity interactions. This paper reports an empirical study on the entity interaction graphs of three systems chosen from different domains: Internet models, molecular dy- namics, and social dynamics, respectively. The results of complex networks analysis on the entity interaction graphs demonstrate that the heterogeneous distribution of connections and highly clustering are universal in these complex systems. Generally, these properties are not obvious at the system modeling stage. Moreover, mutual information theory is used to measure the "principle of persistence" as the predictability of partitioning on multiple processors. This study facilitates better understanding and quantifying of the interaction complexity and provides implications on performance tuning for parallel simulation of large-scale complex systems. Index Terms—entity interaction graph; parallel simulation; empirical study; complex systems; performance analysis
  • Bonan Hou · Yiping Yao ·
    [Show abstract] [Hide abstract]
    ABSTRACT: Efficient large-scale simulation on multiple processors is essential for social dynamics study but still has been proved to be a challenge. Community structure is a ubiquitous property of social networks. It has significant influence on its dynamics and leads the selection of model partition algorithms a critical performance issue. However, the underlying community structure is not well exploited by existing approaches of load-balancing optimizations, which discounted their effectiveness. This paper proposes COMMPAR, a community-based model partitioning approach, which utilizes the community information of social networks for performance tuning. It contains a two-phased network model partitioning as follows: first, community detection algorithm is employed to discover community structure residing in large-scale social networks, second, those communities are further equally partitioned to achieve an appropriate configuration of simulation execution, and facilitates mapping of the communities onto multiple computer processors. Eventually, the experimental results of a random-walk dynamics simulation show that COMMPAR significantly outperforms several existing partitioning approaches, and can efficiently reduce the overhead of interprocessor communications.
    DS-RT '10 Proceedings of the 2010 IEEE/ACM 14th International Symposium on Distributed Simulation and Real Time Applications, Fairfax, Virginia, USA, 17-20 October 2010; 01/2010
  • Bing Wang · Yiping Yao · Bonan Hou · Dongsheng Liao · Dan Chen ·
    [Show abstract] [Hide abstract]
    ABSTRACT: With the pervasive participation from the Chinese netizens in the cyberspace, the Human Flesh Search (HFS) is becoming a phenomenon profoundly affecting the public life in China. While hot debate and discussion arise, it is still not clear how the knowledge fragments are gathered by the netizens, and eventually merge to an answer in a HFS. This paper presents a first study to the knowledge aggregation in HFS. We formally present a normative knowledge aggregation model of the HFS. Simulation experiments have been conducted on several network topologies, with which the convergence feature and the gap between the elites and the crowd in a HFS are reconstructed.
  • Bonan Hou · Yiping Yao · Bing Wang · Dongsheng Liao ·
    [Show abstract] [Hide abstract]
    ABSTRACT: Many social systems can be described in terms of complex networks of interacting dynamic entities. The research on social system's dynamics, such as epidemic or rumor spreading, are often suffered from constraints of relatively small to medium scale networks, or dependence on the underlying network assumptions, resulting in limited capacity for dynamics prediction. This paper proposes SUPE-Net, an efficient parallel simulation environment for large-scale social networks. Built on parallel discrete event simulation engine, the SUPE-Net enables high fidelity modeling of social networks with different topologies and complex dynamics protocols. SUPE-Net automatically maps large-scale social networks onto multi-processors and adopts the discrete event fashion to study the complex network dynamics. Different network topologies, including random, small-world, scale-free, and real networks, can be easily constructed and plugged in for comparative study. The execution efficiency of SUPE-Net is demonstrated by simulation experiments of gossip dynamics on the actor networks with millions of entities.
  • Bonan Hou · Bing Wang · Yiping Yao · Dongsheng Liao · Jifeng Liu ·
    [Show abstract] [Hide abstract]
    ABSTRACT: Computer simulation is a promising approach to better understand the crowd psychological dynamics. The lack of real data in crowd simulations, however, usually results in heavy dependence on assumptions, and thus hampers the confidence. This paper presents our work on the simulation of crowds under confrontation of psychological stimuli. This paper focuses on the psychological aspect of the crowd and introduces the intervention between two different relationship spaces in the crowd. Firstly, with the agent-based approach, psychometrics is employed in the construction of the individual model to improve the trustworthiness. Secondly, we describe in the model the phenomenon that individuals tend to get involved in various relationship spaces, which plays an important role in the procedure of psychological propagation. Finally, a scenario of a crowd in large-scale public activity under terrorist threat is studied, in which the workflow for psychometrics based crowd simulation is demonstrated, and the preliminary results are also discussed.
    Advances in System Simulation, 2009. SIMUL '09. First International Conference on; 10/2009
  • Source
    Bing Wang · Yiping Yao · Yuliang Zhao · Bonan Hou · Shaoliang Peng ·
    [Show abstract] [Hide abstract]
    ABSTRACT: With the increasing demands for large-scale and fine-resolution models, simulations of the reaction-diffusion sys-tems are becoming more and more time consuming. Combined with the Stochastic Simulation Method (SSA), the Parallel Discrete-Event Simulation (PDES) is a promising approach to utilize the parallelism in these models. Since synchronization algorithms play the key role in PDES, in this paper, we experimentally investigate the performance and scalability of optimistic synchronization algorithms in simulations of reaction-diffusion systems. First, the Abstract Next Subvolume Method (ANSM), a variant of the Next Subvolume Method (NSM), is presented. It integrates the logical process (LP) based modeling paradigm with several simulation algorithms including both sequential and parallel execution. Second, based on ANSM, three optimistic synchronization algorithms, including a pure optimistic approach, an optimistic approach with risk-free message sending, and a hybrid approach combined the above two are respectively plugged into the simulation. Third, a group of experiments are conducted to study the characteristics of the synchronization algorithms in the parallel simulation of a typical reaction-diffusion systems. The results show that comparing with the pure optimistic approaches, moderate optimistic approaches are more suitable for the stochastic simulation of reaction-diffusion systems, with respect to both the performance and the scalability.
  • Source
    Bing Wang · Bonan Hou · Yiping Yao · Laibin Yan ·
    [Show abstract] [Hide abstract]
    ABSTRACT: With the development of on-line forum technology and the pervasive participation of the public, the Human Flesh Search is becoming an arising phenomenon which makes a great impact on our daily life. There arose big research interests in social, legal issues resulted from HFS, however, very little work has been conducted to understand how it comes into being and how it dynamically evolves. This paper proposes a modeling and simulation approach incorporating network expansion and GOSSIP propagation with feedback for a better understanding of the human flesh search phenomenon. Based on the acquisition and analysis of the netizens' surfing behavior data, the evolution of the HFS is modeled as a network growth process with proper dynamic input, which is characterized by heavy-tail and burst-oriented distribution, modeling as a Weibulloid process. Then, an improved GOSSIP model with feedback is proposed to represent the information propagation, processing and aggregation during the HFS. New insights for HFS are gained through a set of simulation experiments.
    13th IEEE/ACM International Symposium on Distributed Simulation and Real Time Applications, Singapore, 25-28 October 2009; 01/2009
  • Bonan Hou · Yiping Yao · Bing Wang ·
    [Show abstract] [Hide abstract]
    ABSTRACT: Parallel discrete event simulation (PDES) and high level architecture (HLA) based distributed simulation need to collaborate to conquer the large-scale complex systems. PDES aims at making best use of supercomputing resources while HLA focus on the interoperability of geographically distributed independent simulations. This presents particular challenges on the interoperability between PDES and HLA based simulation. This paper proposes a model mapping framework for mapping the Base Object Model (BOM) conceptual model definition into parallel discrete event simulation models, providing the common semantic basis with federation object models (FOMs) that transformed from the same BOM. With the guide of this framework, the entity types, event types, pattern of interplay and state machines could be mapped into PDES simulation objects, events, event scheduling relationship, and state transition, respectively. A case study about social simulation indicates that this approach not only makes the development of PDES models easy, but also facilitates the conceptual interoperability between PDES and HLA based distributed simulations.
    System Simulation and Scientific Computing, 2008. ICSC 2008. Asia Simulation Conference - 7th International Conference on; 11/2008