基于保守合并策略的复杂网络社区结构发现

为了有效的进行复杂网络社区结构发现,在CNM算法的基础上引进了一种具有额外头部节点的堆数据结构,在社区合并的过程中采用有偏的随机抽样方法,实现了较为保守的合并策略.随机合并算法的输出有更多的机会到达模块度函数的平台区域,在通常情况下其平均表现超过了原始CNM算法.在合成和真实的网络上验证了所提算法的可行性和有效性.     

Community Detection with the Weighted Parsimony Criterion

Community detection in networks has been studied extensively in the last decade. Many criteria, expressing the quality of the partitions obtained, as well as a few exact algorithms and a large number of heuristics have been proposed. The parsimony criterion consists in minimizing the number of edges added or removed from the given network in order to transform it into a set of disjoint cliques.Recently Zhang, Qiu and Zhang have proposed a weighted parsimony model in which a weight coefficient is introduced to balance the numbers of inserted and deleted edges. These authors propose rules to select a good value of the coefficient, use simulated annealing to find optimal or near-optimal solutions and solve a series of real and artificial instances. In the present paper, an algorithm is proposed for solving exactly the weighted parsimony problem for all values of the parameter. This algorithm is based on iteratively solving the problem for a set of given values of the parameter using a row generation algorithm. This procedure is combined with a search procedure to find all lowest breakpoints of the value curve(i.e., the weighted sum of inserted and deleted edges). Computational results on a series of artificial and real world networks from the literature are reported. It appears that several partitions for the same network may be informative and that the set of solutions usually contains at least one intuitively appealing partition.     

复杂网络的社区结构

社区结构作为真实复杂网络所普遍具有的一个重要拓扑特性,在最近10年内得到了广泛而深入的研究.回顾了近几年国内外社区结构研究的主要进展,重点介绍社区发现的研究历程和研究成果,并结合社会计算的背景展望了社区结构研究的未来发展方向和潜在的应用价值.     

用于社群结构探测的小世界特征遗传算法

针对网络社群结构探测问题,探讨一种小世界特征遗传算法.受小世界现象搜索机理和网络结构特征启发,构造一种社群结构局部搜索算子;进而将该算子引入遗传算法基本框架,提出一种小世界特征遗传算法.典型网络的社群结构探测的结果表明,与Girvan-Newman算法和模拟退火算法相比,小世界特征遗传算法探测到的社群结构具有更大的模块性值,从而证明该算法是可用且有效的.     

离散粒子群社区检测算法

针对复杂网络中的社区检测问题,提出了一种基于节点影响力的离散粒子群社区检测方法。该方法以模块度密度作为目标函数,利用离散粒子群算法对其进行优化,在优化过程中提出了节点影响力的概念,其充分利用了网络中节点的相互关系检测网络中的社区结构。同时,在此基础上提出了基于节点影响力的粒子群初始化方法和粒子状态更新方法。利用人工网络数据集和真实网络数据集对所提算法进行测试,实验结果表明,所提算法具有较好的检测结果,能更好地对网络中社区进行划分。     

基于节点相似性的LFM社团发现算法

传统的局部适应度社团发现算法(LFM)在社团结构模糊的网络中精度下降严重。针对此问题,提出LFMJ算法。利用邻居节点信息和改进的杰卡德系数重构网络,使网络结构更为清楚,社团划分结果更为准确。为验证算法,选择了5种算法在LFR网络和真实网络中进行测试,包括LFMJ、LFM和传统的LPA算法以及性能较好的WT和FUA算法。结果表明:在标准LFR网络中,LFMJ精度高于LFM和LPA,与FUA和WT相当;在真实网络和具有重叠结构的LFR网络中,LFMJ精度优于其他4种算法。     

动态社团发现研究综述

为适应当前动态网络数据的发展,对动态网络中的社团结构进行检测、追踪和预测,对国内外关于动态网络社团发现与演化的相关文献进行了综述。归纳了动态网络的社团发现算法,清晰了社团演化事件的定义,并梳理了社团发现与演化算法的应用场景。通过文献梳理,提出将来动态社团的研究应注重在大数据集上的算法优化、在多语境下的信息挖掘和在多场景下的应用性。     

A Link-Based Similarity for Improving Community Detection Based on Label Propagation Algorithm

Community structure is one of the most best-known properties of complex networks. Finding communities help us analyze networks from a mesoscopic viewpoints instead of microscopic or macroscopic one. It helps to understand behavior grouping. Various community detection algorithms have been proposed with some shortcomings in time and space complexity, accuracy, or stability. Label Propagation Algorithm(LPA) is a popular method used for finding communities in an almost-linear time-consuming process. However, its performance is not satisfactory in some metrics such as accuracy and stability. In this paper, a new modified version of LPA is proposed to improve the stability and accuracy of the LPA by defining two concepts-nodes and link strength based on semi-local similarity-,while preserving its simplicity. In the proposed method a new initial node selection strategy, namely the tiebreak strategy, updating order and rule update are presented to solve the random behavior problem of original LPA. The proposed algorithm is evaluated on artificial and real networks. The experiments show that the proposed algorithm is close to linear time complexity with better accuracy than the original LPA and other compared methods. Furthermore, the proposed algorithm has the robustness and stability advantages while the original LPA does not have these features.     

基于节点属性的社群结构探测算法改进

对Vincent D. Blondel等提出的B算法的特点及机理进行了分析, 讨论了节点属性对社群结构探测的可能影响. 进而通过重构初始化网络, 控制节点(社群)合并过程两个方面, 对B算法进行了改进, 获得更优的模块性指标及对应的社群划分. 经计算机模拟网络与实际网络的社群结构探测, 结果表明所提改进算法有效可用, 能在获得较大模块性指标的同时, 获得较好的社群划分结果, 且拥有更低的运算时间.     

二分网络社团结构的比较性定义

为了解决二分网络的聚类问题,借鉴单顶点网络社团结构的比较性定义,提出了直接基于原始二分网络社团结构的比较性定义,这个定义允许社团之间存在重叠,并定义了在二分网络中顶点与社团之间的作用力.在不引进额外参数的情况下,设计了基于此定义的二分网络的聚类算法,并将此算法应用于人工网和一些实际网络中,结果表明这个算法可以比较准确地对网络进行聚类,说明该定义是有效的.     

城市群交通网络层级结构与组团结构识别

为了深入研究城市群交通网络层次结构与组团结构,本文建立复杂网络层级结构和组团结构识别算法,并以长株潭城市群交通网络为例划分城市群交通网络的层级结构与组团结构。通过对长株潭城市群交通网络的层次结构与组团结构的研究可以得出以下几个重要结论:(1)层级结构和组团结构并不是孤立,二者同时存在的。(2)复杂网络的层次结构和组团结构算法能够有效地、准确地划分城市群层级结构与组团结构。(3)河流等自然地理条件对城市交通网络连通性和交通网络中的层次结构和组团结构有至关重要的影响。     

基于非负矩阵分解的复杂网络重构

将网络连边的产生机制和其社团结构结合在一起,基于社团结构决定网络连边的假设推导出节点间的连接概率矩阵并表达为矩阵乘积的形式,然后利用非负矩阵分解得到节点间的连接概率矩阵进行网络重建。设计实验并在几个真实的网络数据上测试,相比基于相似度的网络重构算法,该算法取得了更好的网络重构效果。     

Finding Key Node Sets in Complex Networks Based on Improved Discrete Fireworks Algorithm

Finding out the key node sets that affect network robustness has great practical significance for network protection and network disintegration. In this paper, the problem of finding key node sets in complex networks is defined firstly. Because it is an NP-hard combinatorial optimization problem,discrete fireworks algorithm is introduced to search the optimal solution, which is a swarm intelligence algorithm and is improved by the prior information of networks. To verify the effect of improved discrete fireworks algorithm(IDFA), experiments are carried out on various model networks and real power grid.Results show that the proposed IDFA is obviously superior to the benchmark algorithms, and networks suffer more damage when the key node sets obtained by IDFA are removed from the networks. The key node sets found by IDFA contain a large number of non-central nodes, which provides the authors a new perspective that the seemingly insignificant nodes may also have an important impact on the robustness of the network.     

Scalably revealing the dynamics of soft community structure in complex networks

Revealing the dynamics of community structure is of great concern for scientists from many fields. Specifically, how to quantify the dynamic details of soft community structure is a very interesting topic. In this paper, the authors propose a novel framework to study the scalable dynamic behavior of the soft community structure. First, the authors model the Potts dynamics to detect community structure using a “soft” Markov process. Then the soft stability of in a multiscale view is proposed to naturally uncover the local uniform behavior of spin values across multiple hierarchical levels. Finally, a new partition index is developed to detect fuzzy communities based on the stability and the dynamical information. Experiments on the both synthetically generated and real-world networks verify that the framework can be used to uncover hierarchical community structures effectively and efficiently.     

Orientation in social networks

Stanley Milgram’s small world experiment presents “six degrees of separation” of our world. One phenomenon of the experiment still puzzling us is that how individuals operating with the social network information with their characteristics can be very adept at finding the short chains. The previous works on this issue focus whether on the methods of navigation in a given network structure, or on the effects of additional information to the searching process. In this paper, the authors emphasize that the growth and shape of network architecture is tightly related to the individuals’ attributes. The authors introduce a method to reconstruct nodes’ intimacy degree based on local interaction. Then we provide an intimacy based approach for orientation in networks. The authors find that the basic reason of efficient search in social networks is that the degree of “intimacy” of each pair of nodes decays with the length of their shortest path exponentially. Meanwhile, the model can explain the hubs limitation which was observed in real-world experiment.     

基于超图的超网络:结构及演化机制

为刻画真实数据集的多维关系,以综合视角研究复杂的现实系统,学者将研究重点逐渐从复杂网络转向超网络的研究.超网络分为两类,基于网络的超网络和基于超图的超网络.基于超图的超网络,结构简单,更适合刻画现实世界的多元结构,这种独特优势使其获得越来越多的重视.本文综述并讨论这类超网络的研究进展.首先,界定超网络的概念;其次,归纳用于刻画超网络结构的静态拓扑指标;再次,总结刻画超网络动力学过程的演化模型;并梳理其主要应用领域,最后展望超网络的若干发展方向,旨在推进超网络领域的理论和应用研究.     

基于节点动态连接度的网络社团划分算法

首先定义了节点动态连接度这一概念,然后介绍了基于节点动态连接度的网络社团划分算法,之后再对其中相关参数的取值范围和社团划分结果之间的关系进行了分析,并以Zachary网络为例验证了分析结论。在此基础上,以dolphins、polbooks和football 3个实际网络为对象,进行了社团划分实验,证明了本算法可通过动态调整参数实现对不同网络的社团划分。最后将实验结果与其他几种常见的社团划分算法结果进行了比较,证明了算法的优势,并对算法中需要注意的一些问题进行了说明。     

A New Evaluation Method of Node Importance in Directed Weighted Complex Networks

Current researches on node importance evaluation mainly focus on undirected and unweighted networks, which fail to reflect the real world in a comprehensive and objective way. Based on directed weighted complex network models, the paper introduces the concept of in-weight intensity of nodes and thereby presents a new method to identify key nodes by using an importance evaluation matrix. The method not only considers the direction and weight of edges, but also takes into account the position importance of nodes and the importance contributions of adjacent nodes. Finally, the paper applies the algorithm to a microblog-forwarding network composed of 34 users, then compares the evaluation results with traditional methods. The experiment shows that the method proposed can effectively evaluate the node importance in directed weighted networks.     

变结构离散动态贝叶斯网络及其推理算法

目前的动态贝叶斯网络的研究,是定义在每一个时间片的静态贝叶斯网络结构和参数都一致的基础上,对于过程突变,参数变化等情况就难以适应.为了解决这个问题,提出变结构离散动态贝叶斯网络的概念,并根据概率和动态贝叶斯网络的理论,推导出变结构离散动态贝叶斯网络的推理方法,对算法进行了验证并结合环境变化时的路径选择问题,进行了计算仿真.计算和仿真结果证明了文章提出的变结构离散动态贝叶斯网络的概念和推理算法的正确性.     

时变路网下带时间窗的易腐食品生产-配送问题

为减少易腐食品在生产配送过程中的价值损耗,通常按订单确定的交货期合理组织生产并立即配送.为此,针对多品种易腐食品的集成生产-配送问题,考虑实际配送过程中路网交通的时变特性,建立了以系统总成本最小为目标、带时间窗的易腐食品集成生产-配送问题优化模型.根据该模型特点设计了一种混合遗传算法,并通过数值算例验证了所提模型和算法的有效性.结果表明,路网的时变特性显著影响客户服务水平,优化投入配送的车辆数可有效降低系统总成本和易腐食品的价值损耗.