网络集聚性对节点中心性指标的准确性影响

节点中心性指标是从特定角度对网络某一方面的结构特点进行刻画的度量指标, 因此网络拓扑结构的改变会对节点中心性指标的准确性产生重要影响. 本文利用Holme-Kim模型构建可变集聚系数的无标度网络, 然后采用Susceptible-Infective-Removal模型进行传播影响力的仿真实验, 接着分析了节点中心性指标在不同集聚系数的无标度网络中的准确性. 结果表明, 度中心性和介数中心性的准确性在低集聚系数的网络中表现更好, 特征向量中心性则在高集聚类网络中更准确, 而紧密度中心性的准确性受网络集聚系数的变化影响较小. 因此当网络的集聚系数较低时, 可选择度或者介数作为中心性指标进行网络节点影响力评价; 反之则选择紧密度指标或特征向量指标较好, 尤其当网络的集聚系数接近0.6时特征向量的准确性可以高达到0.85, 是度量小规模网络的较优选择. 另一方面, 传播过程的感染率越高, 度指标和介数指标越可靠, 紧密度和特征向量则相反. 最后Autonomous System实证网络的断边重连实验, 进一步验证了网络集聚性的改变会对节点中心性指标的准确性产生重要影响.     

网络集聚性对节点中心性指标准确性的影响

节点中心性指标是从特定角度对网络某一方面的结构特点进行刻画的度量指标,因此网络拓扑结构的改变会对节点中心性指标的准确性产生重要影响.本文利用Holme-Kim模型构建可变集聚系数的无标度网络,然后采用Susceptible-Infective-Removal模型进行传播影响力的仿真实验,接着分析了节点中心性指标在不同集聚系数的无标度网络中的准确性.结果表明,度中心性和介数中心性的准确性在低集聚系数的网络中表现更好,特征向量中心性则在高集聚类网络中更准确,而紧密度中心性的准确性受网络集聚系数的变化影响较小.因此当网络的集聚系数较低时,可选择度或者介数作为中心性指标进行网络节点影响力评价;反之则选择紧密度指标或特征向量指标较好,尤其当网络的集聚系数接近0.6时特征向量的准确性可以高达到0.85,是度量小规模网络的较优选择.另一方面,传播过程的感染率越高,度指标和介数指标越可靠,紧密度和特征向量则相反.最后Autonomous System实证网络的断边重连实验,进一步验证了网络集聚性的改变会对节点中心性指标的准确性产生重要影响.     

一种改进的基于信息传播率的复杂网络影响力评估算法

评价网络中节点的信息传播影响力对于理解网络结构与网络功能具有重要意义.目前,许多基于最短路径的指标,如接近中心性、介数中心性以及半局部(SP)指标等相继用于评价节点传播影响力.最短路径表示节点间信息传播途径始终选择最优方式,然而实际上网络间的信息传播过程更类似于随机游走,信息的传播途径可以是节点间的任一可达路径,在集聚系数高的网络中,节点的局部高聚簇性有利于信息的有效扩散,若只考虑信息按最优传播方式即最短路径传播,则会低估节点信息传播的能力,从而降低节点影响力的排序精度.综合考虑节点与三步内邻居间的有效可达路径以及信息传播率,提出了一种SP指标的改进算法,即ASP算法.在多个经典的实际网络和人工网络上利用SIR模型对传播过程进行仿真,结果表明ASP指标与度指标、核数指标、接近中心性指标、介数中心性指标以及SP指标相比,可以更精确地对节点传播影响力进行排序.     

微博双向“关注”网络节点中心性及传播 影响力的分析

根据新浪微博的实际数据, 建立了两个基于双向“关注”的用户关系网络, 通过分析网络拓扑统计特征, 发现二者均具有小世界、无标度特征. 通过对节点度、紧密度、介数和k-core 四个网络中心性指标进行实证分析, 发现节点度服从分段幂率分布; 介数相比其他中心性指标差异性最为显著; 两个网络均具有明显的层次性, 但不是所有度值大的节点核数也大; 全局范围内各中心性指标之间存在着较强的相关性, 但在度值较大的节点群这种相关性明显减弱. 此外, 借助基于传染病动力学的SIR信息传播模型来分析四种指标在刻画节点传播能力方面的差异性, 仿真结果表明, 选择具有不同中心性指标的初始传播节点, 对信息传播速度和范围均具有不同影响; 紧密度和k-core较其他指标可以更加准确地描述节点在信息传播中所处的网络核心位置, 这有助于识别信息传播拓扑网络中的关键节点.     

花簇分形无标度网络中节点影响力的区分度

大量研究表明分形尺度特性广泛存在于真实复杂系统中, 且分形结构显著影响网络上的传播动力学行为. 虽然复杂网络的节点传播影响力吸引了越来越多学者的关注, 但依旧缺乏针对分形网络结构的节点影响力的系统研究. 鉴于此, 本文基于花簇分形网络模型, 研究了分形无标度结构上的节点传播影响力. 首先, 对比了不同分形维数下的节点影响力, 结果表明, 当分形维数很小时, 节点影响力的区分度几乎不随节点度变化, 很难区分不同节点的传播影响力, 而随着分形维数的增大, 从全局和局域角度都能很容易识别网络中的超级传播源. 其次, 通过对原分形网络进行不同程度的随机重连来分析网络噪声对节点影响力区分度的影响, 发现在低维分形网络上, 加入网络噪声之后能够容易区分不同节点的影响力, 而在无穷维超分形网络中, 加入网络噪声之后能够区分中间度节点的影响力, 但从全局和局域角度都很难识别中心节点的影响力. 所得结论进一步补充、深化了基于花簇分形网络的节点影响力研究, 研究结果对实际病毒传播的预警控制提供了一定的理论借鉴.     

有限步传播范围期望指标判别节点传播影响力

在线社交网络逐渐成为人们不可或缺的重要工具,识别网络中具有高影响力的节点作为初始传播源,在社会感知与谣言控制等方面具有重要意义.本文基于独立级联模型,给出了一个描述有限步传播范围期望的指标-传播度,并设计了一种高效的递推算法.该指标在局部拓扑结构信息的基础上融合了传播概率对影响力进行刻画,能够较好地反映单个节点的传播影响力.对于多传播源影响力极大化问题,本文提出了一种基于传播度的启发式算法-传播度折扣算法,使得多个传播源的联合影响力最大.最后,将上述方法应用到三个真实网络中,与经典指标和方法相比,该方法不需要知道网络的全局结构信息,而是充分了利用网络的局部结构信息,可以较快地筛选出高传播影响力的传播源.     

基于自规避随机游走的节点排序算法

评估复杂网络系统的节点重要性有助于提升其系统抗毁性和结构稳定性. 目前, 定量节点重要性的排序算法通常基于网络结构的中心性指标如度数、介数、紧密度、特征向量等. 然而, 这些算法需要以知晓网络结构的全局信息为前提, 很难在大规模网络中实际应用. 基于自规避随机游走的思想, 提出一种结合网络结构局域信息和标签扩散的节点排序算法. 该算法综合考虑了节点的直接邻居数量及与其他节点之间的拓扑关系, 能够表征其在复杂网络系统中的结构影响力和重要性. 基于三个典型的实际网络, 通过对极大连通系数、网络谱距离数、节点连边数和脆弱系数等评估指标的实验对比, 结果表明提出的算法显著优于现有的依据局域信息的节点排序算法.     

一种有效的基于三角结构的复杂网络节点影响力度量模型

度量复杂网络中的节点影响力对理解网络的结构和功能起着至关重要的作用.度、介数、紧密度等经典指标能够一定程度上度量节点影响力,k-shell和H-index等指标也可以应用于评价节点影响力.然而这些模型都存在着各自的局限性.本文基于节点与邻居节点之间的三角结构提出了一种有效的节点影响力度量指标模型(local triangle centrality,LTC),该模型不仅考虑节点间的三角结构,同时考虑了周边邻居节点的规模.我们在多个真实复杂网络上进行了大量实验,通过SIR模型进行节点影响力仿真实验,证明LTC指标相比于其他指标能够更加准确地度量节点的传播影响力.节点删除后网络鲁棒性的实验结果也表明LTC指标具有更好效果.     

面向结构洞的复杂网络关键节点排序

复杂网络中的结构洞节点对于信息传播具有重要作用, 现有关键节点排序方法多数没有兼顾结构洞节点和其他类型的关键节点进行排序. 本文根据结构洞理论与关键节点排序相关研究选取了网络约束系数、介数中心性、等级度、效率、网络规模、PageRank值以及聚类系数7个度量指标, 将基于ListNet的排序学习方法引入到复杂网络的关键节点排序问题中, 融合7个度量指标, 构建了一个能够综合评价面向结构洞节点的关键节点排序方法. 采用模拟网络和实际复杂网络进行了大量实验, 人工标准试验结果表明本文排序方法能够综合考虑结构洞节点和核心节点, 关键节点排序与人工排序结果具有较高的一致性. SIR传播模型评估实验结果表明由本文选择TOP-K节点发起的传播能够在较短的传播时间内达到最大的传播范围.     

识别高阶网络传播中最有影响力的节点

识别网络传播中最有影响力的节点是控制传播速度和范围的重要步骤,有助于加速有益信息扩散,抑制流行病、谣言和虚假信息的传播等.已有研究主要基于描述点对交互的低阶复杂网络.然而,现实中个体间的交互不仅发生在点对之间,也发生在3个及以上节点形成的群体中.群体交互可利用高阶网络来刻画,如单纯复形与超图.本文研究单纯复形上最有影响力的传播者识别方法.首先,提出单纯复形上易感-感染-恢复(SIR)微观马尔可夫链方程组,定量刻画单纯复形上的疾病传播动力学.接下来利用微观马尔可夫链方程组计算传播动力学中节点被感染的概率.基于网络结构与传播过程,定义节点的传播中心性,用于排序节点传播影响力.在两类合成单纯复形与4个真实单纯复形上的仿真结果表明,相比于现有高阶网络中心性和复杂网络中最优的中心性指标,本文提出的传播中心性能更准确地识别高阶网络中最有影响力的传播者.     

Crystallographic anisotropy in chemisorption: Nitrogen on tungsten single crystal planes

A molecular beam technique for the determination of sticking probabilities and surface coverages was used in earlier work to investigate the adsorption of nitrogen on tungsten {110}, {111} and {100} single crystal planes. In the present paper these studies have been extended to the {310}, {320} and {411} planes. Absolute sticking probabilities and adatom surface coverages are reported for crystal temperatures between 90 K and 960 K. Crystallographic anisotropy in this system is exemplified by zero coverage sticking probabilities with the crystal at room temperature: {110}, 1̃0^?2; {111}, 0.08; {411}, 0.4; {100}, 0.59; {310}, 0.72; {320}, 0.73. Results for planes on the [001] zone are quantitatively described by a general model developed for adsorption on stepped planes as an extension to the precursor-state order-disorder model for adsorption kinetics of King and Wells. It is shown that nitrogen dissociation only takes place at vacant pairs of {100} sites, but that subsequently the chemisorbed adatoms so formed may migrate out onto {110} terraces. The results are critically analysed in terms of the available LEED and work function data for nitrogen on tungsten single crystal planes, and the general model developed by Adams and Germer.     

Computational complexity of impact size estimation for spreading processes on networks

Spreading processes on networks are often analyzed to understand how the outcome of the process (e.g. the number of affected nodes) depends on structural properties of the underlying network. Most available results are ensemble averages over certain interesting graph classes such as random graphs or graphs with a particular degree distributions. In this paper, we focus instead on determining the expected spreading size and the probability of large spreadings for a single (but arbitrary) given network and study the computational complexity of these problems using reductions from well-known network reliability problems. We show that computing both quantities exactly is intractable, but that the expected spreading size can be efficiently approximated with Monte Carlo sampling. When nodes are weighted to reflect their importance, the problem becomes as hard as the s-t reliability problem, which is not known to yield an efficient randomized approximation scheme up to now. Finally, we give a formal complexity-theoretic argument why there is most likely no randomized constant-factor approximation for the probability of large spreadings, even for the unweighted case. A hybrid Monte Carlo sampling algorithm is proposed that resorts to specialized s-t reliability algorithms for accurately estimating the infection probability of those nodes that are rarely affected by the spreading process.     

Preferential attachment network model with aging and initial attractiveness

In this paper, we generalize the growing network model with preferential attachment for new links to simultaneously include aging and initial attractiveness of nodes. The network evolves with the addition of a new node per unit time, and each new node has m new links that with probability Π_i are connected to nodes i already present in the network. In our model, the preferential attachment probability Π_i is proportional not only to k_i + A, the sum of the old node i's degree k_i and its initial attractiveness A, but also to the aging factor ${\tau }_{i}^{-\alpha }$, where τ_i is the age of the old node i. That is, ${{\rm{\Pi }}}_{i}\propto ({k}_{i}+A){\tau }_{i}^{-\alpha }$. Based on the continuum approximation, we present a mean-field analysis that predicts the degree dynamics of the network structure. We show that depending on the aging parameter α two different network topologies can emerge. For α < 1, the network exhibits scaling behavior with a power-law degree distribution P(k) ∝ k^−γ for large k where the scaling exponent γ increases with the aging parameter α and is linearly correlated with the ratio A/m. Moreover, the average degree k(t_i, t) at time t for any node i that is added into the network at time t_i scales as $k({t}_{i},t)\propto {t}_{i}^{-\beta }$ where 1/β is a linear function of A/m. For α > 1, such scaling behavior disappears and the degree distribution is exponential.     

复杂网络中最小K-核节点的传播能力分析

K-核分解方法对于识别复杂网络传播动力学中最重要节点具有重要的价值, 然而该方法无法对复杂网络中大量最小K-核节点的传播能力进行准确度量. 本文主要考察最小K-核节点的传播行为, 利用其邻居的K-核信息, 提出一种度量这类节点传播能力的方法. 实证网络数据集的传播行为仿真结果表明, 该方法与度、介数等指标相比更能准确度量最小K-核节点的传播能力. 关键词： 复杂网络 传播能力 K-核分解 最小K-核节点     

Ranking the spreading influence of nodes in complex networks: An extended weighted degree centrality based on a remaining minimum degree decomposition

Ranking the spreading influence of nodes is crucial for developing strategies to control the spreading process on complex networks. In this letter, we define, for the first time, a remaining minimum degree (RMD) decomposition by removing the node(s) with the minimum degree iteratively. Based on the RMD decomposition, a weighted degree (WD) is presented by utilizing the RMD indices of the nearest neighbors of a node. WD assigns a weight to each degree of this node, which can distinguish the contribution of each degree to the spreading influence. Further, an extended weighted degree (EWD) centrality is proposed by extending the WD of the nearest neighbors of a node. Assuming that the spreading process on networks follows the Susceptible-Infectious-Recovered (SIR) model, we perform extensive experiments on a series of synthetic and real networks to comprehensively evaluate the performance of EWD and other eleven representative measures. The experimental results show that EWD is a relatively efficient measure in running efficiency, it exposes an advantage in accuracy in the networks with a relatively small degree heterogeneity, as well as exposes a competitive performance in resolution.     

钼单晶体的范性形变

在本文中,用金相和X射线方法,研究了24个取向不同的钼单晶体在-80℃(-50℃)、27℃、1000℃和～2000℃拉伸后的情况。分析研究的结果,认为观察到的{112}、{123}、{145}等滑移痕迹,是由于在两组不平行的{110}面上,沿着同一个<111>方向组合滑移后构成的外观面貌,而滑移面是密排的{110}面。外观滑移面(从滑移痕迹测定出的)会随样品取向不同而发生变化。当变形温度改变时,同一个样品的外观滑移面可能改变,也可能不改变,这要由样品的取向来决定。 关键词：     

基于区域密度曲线识别网络上的多影响力节点

复杂网络多影响力节点的识别可以帮助理解网络的结构和功能,具有重要的理论意义和应用价值.本文提出一种基于网络区域密度曲线的多影响力节点的识别方法.应用两种不同的传播模型,在不同网络上与其他中心性指标进行了比较.结果表明,基于区域密度曲线的识别方法能够更好地识别网络中的多影响力节点,选中的影响力节点之间的分布较为分散,自身也比较重要.本文所提方法是基于网络的局部信息,计算的时间复杂度较低.     

Harmonic measure for percolation and ising clusters including rare events

We obtain the harmonic measure of the hulls of critical percolation clusters and Ising-model Fortuin-Kastelyn clusters using a biased random-walk sampling technique which allows us to measure probabilities as small as 10{-300}. We find the multifractal D(q) spectrum including regions of small and negative q. Our results for external hulls agree with Duplantier's theoretical predictions for D(q) and his exponent -23/24 for the harmonic measure probability distribution for percolation. For the complete hull, we find the probability decays with an exponent of -1 for both systems.     

Optical transitions and upconversion properties of Er^3＋-doped chloride tellurite glasses

Er^{3+}-doped lead chloride tellurite glasses were prepared using the conventional melting and quenching method. The absorption spectra were measured and the Judd－Ofelt analysis was performed. The spectroscopic parameters such as the intensity parameters, transition probabilities, radiative lifetimes, and branching ratios were obtained. Intense infrared emission and visible upconversion luminescence under 976 nm excitation were observed. For the 1.55μm emission, the full width at half maximum and the emission cross sections are more than 50 nm and 8×10^{- 20}cm^2, respectively. Three efficient visible luminescences centred at 525, 547, and 658nm are assigned to the transitions from the excited states {}^{2}H_{11/2}, {}^{4}S_{3/2}, and {}^{4}F_{9/2} to the ground state {}^{4}I_{15/2}, respectively. The upconversion mechanisms and the power-dependent intensities are also discussed and evaluated.