网络的模块矩阵及其社团结构指标

探讨了复杂网络的模块矩阵的正(负)特征谱与网络的社团结构(反社团结构)的关系,给出了反映网络社团结构性质的相关定义.利用模块矩阵的多个特征值与特征向量,引入反映个体对所处社团的依附程度一种结构中心化指标.利用人工网络与实际网络数据,将这种指标与几种经典的中心化指标进行了比较.结果表明该指标具有较好的分辨率并与度指标具有一定程度的相关性.     

基于联合矩阵分解的节点多属性网络社团检测

发现复杂网络中的社团结构在社会网络、生物组织网络和在线网络等复杂网络中具备十分重要的意义. 针对社交媒体网络的社团检测通常需要利用两种信息源: 网络拓扑结构特征和节点属性特征, 丰富的节点内容属性信息为社团检测的增加了灵活性和挑战. 传统方法是要么仅针对这两者信息之一进行单独挖掘, 或者将两者信息得到的社团结果进行线性叠加判决, 不能有效进行信息源的融合. 本文将节点的多维属性特征作为社团划分的一种有效协同学习项进行研究, 将两者信息源进行融合分析, 提出了一种基于联合矩阵分解的节点多属性网络社团检测算法CDJMF, 提高了社团检测的有效性和鲁棒性. 实验表明, 本文所提的方法能够有效利用节点的属性信息指导社团检测, 具备更高的社团划分质量.     

复杂网络中社团结构发现的多分辨率密度模块度

现实中的许多复杂网络呈现出明显的模块性或社团性.模块度是衡量社团结构划分优劣的效益函数, 它也通常被用作社团结构探测的目标函数,但最为广泛使用的Newman-Girvan模块度却存在着分辨率限制问题,多分辨率模块度也不能克服误合并社团和误分裂社团同时存在的缺陷. 本文在网络密度的基础上提出了多分辨率的密度模块度函数, 通过实验和分析证实了该函数能够使社团结构的误划分率显著降低, 而且能够体现出网络社团结构是一个有机整体,不是各个社团的简单相加.     

基于函数矩阵的一类混沌系统同步

研究了一类混沌系统的同步问题、基于稳定性理论和极点配置技术,设计了两个混沌系统之间的同步方案,实现两个混沌系统之间的同步,通过函数矩阵,实现驱动系统和响应系统的状态变量按给定的函数矩阵同步,同时证明了该方法同样适用于两个混沌系统之间的滞后同步,通过对Lorenz混沌系统和Lorenz超混沌系统的数值模拟,进一步验证了所提方案的有效性。     

Community detection based on modularity and an improved genetic algorithm

Complex networks are widely applied in every aspect of human society, and community detection is a research hotspot in complex networks. Many algorithms use modularity as the objective function, which can simplify the algorithm. In this paper, a community detection method based on modularity and an improved genetic algorithm (MIGA) is put forward. MIGA takes the modularity Q$Q$ as the objective function, which can simplify the algorithm, and uses prior information (the number of community structures), which makes the algorithm more targeted and improves the stability and accuracy of community detection. Meanwhile, MIGA takes the simulated annealing method as the local search method, which can improve the ability of local search by adjusting the parameters. Compared with the state-of-art algorithms, simulation results on computer-generated and four real-world networks reflect the effectiveness of MIGA.     

Quantum inspired evolutionary algorithm for community detection in complex networks

Community structure is indispensable to discover the potential property of complex network systems. In this paper we propose two algorithms (QIEA-net and iQIEA-net) to discover communities in social networks by optimizing modularity. Unlike many existing methods, the proposed algorithms adopt quantum inspired evolutionary algorithm (QIEA) to optimize a population of solutions and do not need to give the number of community beforehand, which is determined by optimizing the value of modularity function and needs no human intervention. In order to accelerate the convergence speed, in iQIEA-net, we apply the result of classical partitioning algorithm as a guiding quantum individual, which can instruct other quantum individuals' evolution. We demonstrate the potential of two algorithms on five real social networks. The results of comparison with other community detection algorithms prove our approaches have very competitive performance.     

Efficient overlapping community detection in huge real-world networks

The detection of overlapping community structure in networks can give insight into the structures and functions of many complex systems. In this paper, we propose a simple but efficient overlapping community detection method for very large real-world networks. Taking a high-quality, non-overlapping partition generated by existing, efficient, non-overlapping community detection methods as input, our method identifies overlapping nodes between each pair of connected non-overlapping communities in turn. Through our analysis on modularity, we deduce that, to become an overlapping node without demolishing modularity, nodes should satisfy a specific condition presented in this paper. The proposed algorithm outputs high quality overlapping communities by efficiently identifying overlapping nodes that satisfy the above condition. Experiments on synthetic and real-world networks show that in most cases our method is better than other algorithms either in the quality of results or the computational performance. In some cases, our method is the only one that can produce overlapping communities in the very large real-world networks used in the experiments.     

Synchronization on overlapping community network

In this paper, we propose a simple random network model with overlapping communities controlled by several parameters, and investigate the influence of the overlapping community structure on the synchronization behavior under different parameters. It is found that the synchronizability of the network is mainly influenced by the overlapping size of the communities and the connectivity density of the overlapped group to the other interrelated communities, and has nothing to do with the intra-connectivity of the overlapped group. In addition, it is found that the highly interconnected communities can be almost synchronized in a given time scale, whereas the overlapped group is far from synchronization. Furthermore, the instantaneous frequencies of the nodes in the communities and their overlapped group are also investigated, which show that the nodes in the overlapped group will exhibit a remarkable oscillation with a weighted mean frequency of the other correlative communities.     

Synchronization speed of identical oscillators on community networks

Synchronizability of complex oscillators networks has attracted much research interest in recent years. In contrast, in this paper we investigate numerically the synchronization speed, rather than the synchronizability or synchronization stability, of identical oscillators on complex networks with communities. A new weighted community network model is employed here, in which the community strength could be tunable by one parameter δ. The results showed that the synchronization speed of identical oscillators on community networks could reach a maximal value when δ is around 0.1. We argue that this is induced by the competition between the community partition and the scale-free property of the networks. Moreover, we have given the corresponding analysis through the second least eigenvalue λ₂ of the Laplacian matrix of the network which supports the previous result that the synchronization speed is determined by the value of λ₂.     

基于混沌神经网络的盲检测改进新算法

针对Hopfield神经网络的多起点问题,提出了一种新的基于混沌神经网络的盲信号检测算法,实现了二进制移相键控信号盲检测.据此进一步提出双sigmoid混沌神经网络模型,构造了新的能量函数,且证明了该模型的稳定性,并对网络参数进行配置.仿真实验表明:混沌神经网络能够避免局部极小点且具备较强的抗噪性能,双sigmoid混沌神经网络则继承了其所有的优点,且其收敛速度更快,仅需更短的接收数据即可到达全局真实平衡点,从而降低了算法的计算复杂度,减少了运行时间.     

An efficient community detection method based on rank centrality

Community detection is a very important problem in social network analysis. Classical clustering approach, K$K$-means, has been shown to be very efficient to detect communities in networks. However, K$K$-means is quite sensitive to the initial centroids or seeds, especially when it is used to detect communities. To solve this problem, in this study, we propose an efficient algorithm K$K$-rank, which selects the top-K$K$ nodes with the highest rank centrality as the initial seeds, and updates these seeds by using an iterative technique like K$K$-means. Then we extend K$K$-rank to partition directed, weighted networks, and to detect overlapping communities. The empirical study on synthetic and real networks show that K$K$-rank is robust and better than the state-of-the-art algorithms including K$K$-means, BGLL, LPA, infomap and OSLOM.     

基于演化博弈论的行人与机动车冲突演化机理研究

行人与机动车冲突时,各自都会在经过简单判断后以一定的概率选择通过.本文根据人车冲突的实际情景提出基础收益、冲突损失、等待损失以及互让损失的概念,据此构建行人与机动车的冲突博弈矩阵,并依据演化分析范式,建立人车冲突演化的动力学模型.对不同交通情形下均衡点的位置、稳定性以及系统演化机理进行深入分析,发现不同的行人与机动车的冲突损失和等待损失相对大小,对应系统的演化方向不同,可能的演化方向包括"人让车","车让人","人让车,同时车让人"以及"人不让车,车不让人".此外,定义机会损失的交通概念,据此分析系统关于行人与机动车的互让损失以及机会损失的灵敏度,发现行人或机动车互让损失的增加对于各自通过概率有着上升促进和下降抑制作用,而机会损失的作用恰好与互让损失相反.本文建立的动力学模型可以为人车冲突演化方向的宏观调控提供理论依据.     

基于簇相似度的网络社团结构探测算法

社团结构对复杂系统的结构特性和动力学特性有重要影响.提出了一个度量社团相似度的模型,称为簇相似度.该模型能够度量两个社团的相似度大小,为研究社团间的作用机制提供帮助.而且基于该模型,设计了一个社团划分算法.算法采用层次聚类的思想,每次合并两个相似度最大的社团,并通过一个评价函数选择最优社团划分.数值实验以及与CNM,GN,EigenMod等主流算法做比较,表明本算法的精度和效率都比较高,尤其对于边密度较高的网络,性能非常理想.     

Synchronization of coupled logistic maps on random community networks

The collective synchronization of a system of coupled logistic maps on random community networks is investigated. It is found that the synchronizability of the community network is affected by two factors when the size of the network and the number of connections are fixed. One is the number of communities denoted by the parameter rn, and the other is the ratio σ of the connection probability p of each pair of nodes within each community to the connection probability q of each pair of nodes among different communities. Theoretical analysis and numerical results indicate that larger rn and smaller σ are the key to the enhancement of network synchronizability. We also testify synchronous properties of the system by analysing the largest Lyapunov exponents of the system.     

基于演化算法求解电子电荷量e的值

应用演化算法的智能寻优直接求出一组有误差油滴电荷量的最大公约数,得到了电子电荷量e的值．     

Community detection in signed networks based on discrete-time model

Community detection in signed networks has been studied widely in recent years. In this paper, a discrete difference equation is proposed to imitate the consistently changing phases of the nodes. During the interaction, each node will update its phase based on the difference equation. Each node has many different nodes connected with it, and these neighbors have different influences on it. The similarity between two nodes is applied to describe the influences between them. Nodes with high positive similarities will get together and nodes with negative similarities will be far away from each other.Communities are detected ultimately when the phases of the nodes are stable. Experiments on real world and synthetic signed networks show the efficiency of detection performance. Moreover, the presented method gains better detection performance than two existing good algorithms.     

Phase synchronization on small-world networks with community structure

In this paper, we propose a simple model that can generate small-world network with community structure. The network is introduced as a tunable community organization with parameter r, which is directly measured by the ratio of inter- to intra-community connectivity, and a smaller r corresponds to a stronger community structure. The structure properties, including the degree distribution, clustering, the communication efficiency and modularity are also analysed for the network. In addition, by using the Kuramoto model, we investigated the phase synchronization on this network, and found that increasing the fuzziness of community structure will markedly enhance the network synchronizability; however, in an abnormal region (r ≤ 0.001), the network has even worse synchronizability than the case of isolated communities (r = 0). Furthermore, this network exhibits a remarkable synchronization behaviour in topological scales: the oscillators of high densely interconnected communities synchronize more easily, and more rapidly than the whole network.     

Detecting local community structures in complex networks based on local degree central nodes

Detecting local communities in real-world graphs such as large social networks, web graphs, and biological networks has received a great deal of attention because obtaining complete information from a large network is still difficult and unrealistic nowadays. In this paper, we define the term local degree central node whose degree is greater than or equal to the degree of its neighbor nodes. A new method based on the local degree central node to detect the local community is proposed. In our method, the local community is not discovered from the given starting node, but from the local degree central node that is associated with the given starting node. Experiments show that the local central nodes are key nodes of communities in complex networks and the local communities detected by our method have high accuracy. Our algorithm can discover local communities accurately for more nodes and is an effective method to explore community structures of large networks.     

Growth curve based label propagation algorithm for community detection

How to better and faster identify the community structure is a hot issue in complex networks. During the past decades, various attempts have been made to solve this issue. Amongst them, without doubt, label propagation algorithm (LPA) is one of the most satisfying answers, especially for large-scale networks. However, it has one major flaw that when the community structure is not clear enough, a monster community tends to form. To address this issue, we set a growth curve for communities, gradually increasing from a low capacity to a higher capacity over time. Further, we improve the mechanism of label choosing for small communities to escape from local maximum. The experimental results on both synthetic and real networks demonstrate that our algorithm not only enhances the detection ability of the traditional label propagation algorithm, but also improves the quality of the identified communities.     

一种基于外差探测的光纤Bragg光栅温度传感器

提出了一种基于光纤Bragg光栅的温度传感器，阐述了光纤Bragg光栅的温度传感机理，用2个相同的光纤Bragg光栅构成折叠式Mach Zehnder(M Z)干涉仪，其中一个光栅作为参考臂，另一个作为传感臂：采用外差探测技术来测量外界的温度物理量。当温度发生变化，Bragg光栅的波长也随之改变。外差探测用来探测传感臂和参考臂由于温度变化引起的输出信号的频率差异。对其动态测量范围和灵敏度也进行了分析。