基于联合矩阵分解的动态异质网络社区发现方法

动态网络的社区发现是目前复杂网络分析领域的重要研究内容,然而现有动态网络社区发现方法主要针对同质网络,当网络包含多种异质信息时,现有方法不再适用。针对这个问题,本文提出了一个基于联合矩阵分解的动态异质网络社区发现方法,首先计算动态异质网路中各个快照图的拓扑相似度矩阵和多关系相似度矩阵,其次利用时序联合非负矩阵分解方法,约束各个时刻快照图的社区划分,最后在真实网络数据集上的实验结果表明,该算法可以有效检测出动态异质网络中潜在的社区结构。     

A novel parallel community detection scheme based on label propagation

Community detection is one of the most important ways to reflect the structures and mechanisms of a social network. The overlapping communities are more in line with the reality of the social networks. In society, the phenomenon of some members sharing memberships among different communities reflects as overlapping communities in the networks. Dealing with big data networks, it is a challenging and computationally complex problem to detect overlapping communities. In this paper, we propose highly scalable variants of a community-detection algorithm in a parallel manner called Label Propagation with nodes Confidence (PLPAC). We introduce MapReduce into our scheme to process the big data in a parallel manner and guarantee the efficiency of community detection. We implemented the algorithm on artificial networks as well as real networks to evaluate the accuracy and speedup of the proposed method. Experimental results on datasets from different scenarios illustrate that the improved label propagation method outperforms the state-of-the-art methods in terms of accuracy and time efficiency.     

模块度增量与局部模块度引导下的社区发现算法

社区结构是复杂网络的重要特性之一,基于层次聚类的社区发现算法很好地利用了模块度来挖掘网络中的社区结构,但其局限性也导致算法对社区结构复杂的网络划分不够准确、无法发现小于一定规模的社区。在层次聚类的基础上,提出引入局部模块度来弥补模块度在划分社区时的不足,避免可能出现的划分不合理情况。通过真实数据集和人工网络进行了验证,实验结果证明,该算法具有可行性与有效性。     

结合拓扑势与信任度调整的重叠社区发现算法

现实世界中的复杂系统可建模为复杂网络,探究复杂网络中的社区发现算法对于分析复杂网络的拓扑结构和层次结构具有重要作用。早期研究通常将网络中的节点局限在一个社区中,但随着研究的深入发现社区结构呈现重叠特性。针对现有重叠社区发现算法存在划分社区结构不稳定、忽略节点交互和属性等问题,提出一种基于网络拓扑势与信任度调整的重叠社区发现算法。融合节点的属性和结构特征计算节点的拓扑势,依据节点的拓扑势选取核心节点。从核心节点出发构建初始社区群,计算各个社区间的调整信任度,实现社区的合并与再调整,从而识别重叠社区。在多个人工模拟网络和真实网络数据集上的实验结果表明,与基于贪婪派系扩张、种子扩张等的重叠社区发现算法相比,该算法将扩展模块度最高提升至0.719,能有效识别社区结构及重叠节点,提升重叠社区检测性能。     

一种基于粗糙集和密度峰值的重叠社区发现方法

随着互联网和社会的发展,各个领域每天都会产生大量相互关联、彼此依赖的数据,这些数据根据不同的主题形成了各种复杂网络。挖掘社区结构是复杂网络领域中的一项重要研究内容,因为其在推荐系统、行为预测和信息传播等方面具有极其重要的意义。社区结构中的重叠社区结构在生活中普遍存在,更具有实际研究意义。为有效发现复杂网络中的重叠社区,文中引入了粗糙集理论对社区进行分析,识别出重叠节点,进而提出了一种基于粗糙集和密度峰值的重叠社区发现方法OCDRD(Overlapping Community Detection Algorithm Based on Rough Sets and Density Peaks)。该方法在传统网络节点局部相似性度量的基础上,结合灰色关联分析方法求出网络节点间的全局相似性,进而将其转化为节点间距离。将密度峰值聚类算法的思想应用于该算法中,以根据网络结构自动选取社区中心节点。依据网络中节点的距离比例关系,定义了社区的上近似、下近似以及边界域。最后,不断调整距离比率阈值并进行划分迭代,在每次迭代中针对社区的边界域进行计算,从而获得最佳重叠社区划分结构。在LFR基准人工网络数据集和真实网络数据集上,基于标准互信息(Normalized Mutual Information,NMI)和具有重叠性模块度EQ这两个评价指标,将OCDRD方法与近几年效果较好的其他社区发现算法进行测试比较。实验结果显示,OCDRD方法在社区划分结构方面整体优于其他社区发现算法,表明了该算法的可行性和有效性。     

Finding overlapping communities in a complex network of social linkages and Internet of things

Online social networks play an important role in today’s Internet. These social networks contain huge amounts of data and the integrated framework of SN with Internet of things (IoT) presents a challenging problem. IoT is the ubiquitous interconnection of everyday items of interest (things), providing connectivity anytime, anywhere, and with anything. Like biological, co-authorship, and virus-spread networks, IoT and Social Network (SN) can be characterized to be complex networks containing substantial useful information. In the past few years, community detection in graphs has been an active area of research (Lee and Won in Proceedings of IEEE SoutheastCon, pp. 1–5, 2012). Many graph mining algorithms have been proposed, but none of them can help in capturing an important dimension of SNs, which is friendship. A friend circle expands with the help of mutual friends, and, thus, mutual friends play an important role in social networks’ growth. We propose two graph clustering algorithms: one for undirected graphs such as Facebook and Google+, and the other for directed graphs such as Twitter. The algorithms extract communities, and based on the access control policy nodes share resources (things). In the proposed Community Detection in Integrated IoT and SN (CDIISN) algorithm, we divide the nodes/actors of complex networks into basic, and IoT nodes. We, then, execute the community detection algorithm on them. We take nodes of a graph as members of a SN, and edges depicting the relations between the nodes. The CDIISN algorithm is purely deterministic, and no fuzzy communities are formed. It is known that one community detection algorithm is not suitable for all types of networks. For different network structures, different algorithms exhibit different results, and methods of execution. However, in our proposed method, the community detection algorithm can be modified as desired by a user based on the network connections. The proposed community detection approach is unique in the sense that a user can define his community detection criteria based on the kind of network.     

基于节点依赖度和相似社团融合的社团结构发现算法

社团结构作为复杂网络的拓扑特性之一具有重要的理论和实践意义。提出一种基于节点依赖度和相似社团融合的社团结构发现算法,首先根据依赖度和相似度的定义将整个网络划分成若干个平均集聚系数较大的局部网络,构成网络的基础骨架社团;然后根据连接度的定义不断将社团边缘的节点和小社团吸收到相应的骨架网络中去,直到所有节点都得到准确的社团划分。算法在Zachary空手道俱乐部网络和海豚社会网络中进行了社团划分实验,并与GN算法和Newman快速算法进行了比较,结果表明该算法可以有效地划分社团边缘的模糊节点,社团划分结果具有较高的准确度。     

A hybrid artificial immune network for detecting communities in complex networks

One of the challenging problems when studying complex networks is the detection of sub-structures, called communities. Network communities emerge as dense parts, while they may have a few relationships to each other. Indeed, communities are latent among a mass of nodes and edges in a sparse network. This characteristic makes the community detection process more difficult. Among community detection approaches, modularity maximization has attracted much attention in recent years. In this paper, modularity density (D value) has been employed to discover real community structures. Due to the inadequacy of previous mathematical models in finding the correct number of communities, this paper first formulates a mixed integer non-linear program to detect communities without any need of prior knowledge about their number. Moreover, the mathematical models often suffer from NP-Hardness. In order to overcome this limitation, a new hybrid artificial immune network (HAIN) has been proposed in this paper. HAIN aims to use a network’s properties in an efficient way. To do so, this algorithm employs major components of the pure artificial immune network, hybridized with a well-known heuristic, to provide a powerful and parallel search mechanism. The combination of cloning and affinity maturation components, a strong local search routine, and the presence of network suppression and diversity are the main components. The experimental results on artificial and real-world complex networks illustrate that the proposed community detection algorithm provides a useful paradigm for robustly discovering community structures.     

Incremental community miner for dynamic networks

Human relationships have led to complex communication networks among different individuals in a society. As the nature of relationship is change, these networks will change over the time too which makes them dynamic networks including several consecutive snapshots. Nowadays, the pervasiveness of electronic communication networks, so called Social Networks, has facilitated obtaining this valuable communication information and highlighted as one of the most interesting researchers in the field of data mining, called social network mining. One of the most challenging issues in the field of social network mining is community detection. It means to detect hidden communities in a social network based on the available information. This study proposes an appropriate solution to find and track communities in a dynamic social network based on the local information. Our approach tries to detect communities by finding initial kernels and maintaining them in the next snapshots. Using well-known datasets, the investigation and comparison of the proposed method with some state-of-the-art approaches indicates that the performance and computation complexity of our method is promising and can outperform its competitors.     

一种基于增量式谱聚类的动态社区自适应发现算法

针对当前复杂网络动态社区发现的热点问题, 提出一种面向静态网络社区发现的链接相关线性谱聚类算法, 并在此基础上提出一种基于增量式谱聚类的动态社区自适应发现算法. 动态社区发现算法引入归一化图形拉普拉斯矩阵呈现复杂网络节点之间的关 系,采用拉普拉斯本征映射将节点投影到k维欧式空间.为解决离群节点影响谱聚类的效果和启发式确定复杂网络社区数量的问题, 利用提出的链接相关线性谱聚类算法发现初始时间片的社区结构, 使发现社区的过程能够以较低的时间开销自适应地挖掘复杂网络社区结构. 此后, 对于后续相邻的时间片, 提出的增量式谱聚类算法以前一时间片聚类获得的社区特征为基础, 通过调整链接相关线性谱聚类算法实现对后一时间片的增量聚类, 以达到自适应地发现复杂网络动态社区的目的. 在多个数据集的实验表明, 提出的链接相关线性谱聚类算法能够有效地检测出复杂网络中的社区结构以及基于 增量式谱聚类的动态社区自适应发现算法能够有效地挖掘网络中动态社区的演化过程.     

基于标签传递的有向图重叠社团发现算法研究

社团发现作为网络科学中一个重要的基础问题受到了广泛的关注和重视.针对社团结构的研究为我们提供了从中尺度上分析和理解网络的途径,具有重要的理论和实际意义.已有的研究大多关注无向图和非重叠社团的发现.本文基于标签传递和用户排序的思想设计了一个有向图上的重叠社团发现算法,实际数据上的实验表明了算法在发现用户多重社团属性和确定社团规模方面的有效性.     

局部搜索与遗传算法结合的大规模复杂网络社区探测

基于遗传算法的复杂网络社区探测是当前的研究热点. 针对该问题,本文在分析网络模块性函数Q的局部单调性的基础上, 给出一种快速、有效的局部搜索变异策略, 同时为兼顾初始种群的精度和多样性以达到进一步提高搜索效率的目的, 采用了标签传播作为初始种群的产生方法;综上,提出了一个结合局部搜索的遗传算法(Genetic algorithm with local search, LGA). 在基准网络及大规模复杂网络上对LGA进行测试, 并与当前具有代表性的社区探测算法进行比较, 实验结果表明了文中算法的有效性与高效性.     

Identifying different community members in complex networks based on topology potential

There has been considerable interest in designing algorithms for detecting community structure in real-world complex networks. A majority of these algorithms assume that communities are disjoint, placing each vertex in only one cluster. However, in nature, it is a matter of common experience that communities often overlap and members often play multiple roles in a network topology. To further investigate these properties of overlapping communities and heterogeneity within the network topology, a new method is proposed to divide networks into separate communities by spreading outward from each local important element and extracting its neighbors within the same group in each spreading operation. When compared with the state of the art, our new algorithm can not only classify different types of nodes at a more fine-grained scale successfully but also detect community structure more effectively. We also evaluate our algorithm using the standard data sets. Our results show that it performed well not only in the efficiency of algorithm, but also with a higher accuracy of partition results.     

动态加权网络中的演化社区发现算法研究

在动态网络中发现社区结构是一个非常复杂而有意义的过程,可以更好地观察和分析网络的演化情况。针对动态加权网络中的社区发现问题,提出了一种结合历史网络社区结构的算法,叫做动态加权网络中的演化社区发现算法（ECDA）。该算法分为两步：结合历史社区和网络结构信息,计算当前时间跳的输入矩阵;然后通过该输入矩阵计算得到结合历史时间跳信息的社区划分结果。该算法有以下优点：可以自动发现动态加权网络中每个时间跳的社区结构;对网络结构的变化和社区结构的变化具有较高的敏锐性。在人工数据集和真实数据集中进行了实验,实验结果证明该算法可以有效地发现动态加权网络中的社区结构,与其他算法相比具有较好的竞争力。     

一种基于信息扩散的复杂网络重叠社区发现算法

社区结构可以为网络的其他分析挖掘提供中观尺度的分析视角,在大规模复杂网络的各项研究中是一项非常重要而基础的工作。社区的重叠是真实世界网络中常见的一种现象,重叠社区结构可以更准确地描述网络中真实的结构信息,因此,复杂网络重叠社区发现具有更加突出的现实意义。在综合对比分析了当前主要的重叠社区发现算法的基础上,结合信息论的相关知识,给出了一种基于信息论的社区定义,并进一步借鉴信息传播理论,从单个节点对关于某种主题的信息的掌握程度的角度出发提出了一种复杂网络重叠社区结构发现算法。基于实际数据集的相关实验表明,与传统的社区定义和社区发现算法相比,本算法发现的重叠社区从内容角度来看具有更加明确的实际意义,并且具有较低的时间复杂度。     

基于局部相似性的复杂网络社区发现方法

复杂网络是复杂系统的典型表现形式, 社区结构是复杂网络最重要的结构特征之一. 针对复杂网络的社区结构发现问题, 本文提出一种新的局部相似性度量, 并结合层次聚类算法用于社区结构发现. 相对全局的相似性度量, 本文提出的相似性度量具有较低的计算开销; 同时又能很好地刻画网络的结构特征, 克服了传统局部相似性度量在某些情形下对节点相似性的低估倾向. 为了将局部相似性度量用于社区结构发现, 推广了传统的Ward层次聚类算法, 使之适用于具有相似性度量的任意对象, 并将其用于复杂网络社区结构发现. 在合成和真实世界的网络上进行了实验, 并与典型算法进行了比较, 实验结果表明所提算法的可行性和有效性.     

LinkLPA: A Link‐Based Label Propagation Algorithm for Overlapping Community Detection in Networks

Community detection is an important methodology for understanding the intrinsic structure and function of complex networks. Because overlapping community is one of the characteristics of real‐world networks and should be considered for community detection, in this article, we propose an algorithm, called link‐based label propagation algorithm (LinkLPA), to detect overlapping communities. Because the link partition is conceptually natural for the problem of overlapping community detection, LinkLPA first transforms node partition problem into link partition problem and employs a new label propagation algorithm with preference on links instead of nodes to detect communities due to the simplicity and efficiency of label propagation algorithm. Then the proposed LinkLPA performs a postprocessing to refine the detected overlapping communities by avoiding over‐overlapping and incorrect partition of weak ties. Experimental results on a large number of real‐world and synthetic networks show that the proposed method achieves high accuracy on detecting overlapping communities in networks.     

基于蚁群优化的二分网络社区挖掘

近年来,网络社区挖掘得到了极大的关注,尤其是针对二分网络的社区挖掘。二分网络社区挖掘对于研究复杂网络有非常重要的理论意义和实用价值。提出了一个基于蚁群优化的二分网络社区挖掘算法。该算法首先将二分网络社区挖掘问题转化成一个优化问题,建立一个可供蚂蚁搜索的图模型。同时,根据顶点的拓扑结构定义启发式信息。每只蚂蚁根据每条路径上的信息素和启发式信息选择路径,构造出一个社区的划分,再用二分模块度去衡量社区划分的优劣。实验结果表明,该算法不但可以较准确地识别二分网络的社区数。而且可以获得高质量的社区划分。     

Discovery of extreme events-related communities in contrasting groups of physical system networks

The latent behavior of a physical system that can exhibit extreme events such as hurricanes or rainfalls, is complex. Recently, a very promising means for studying complex systems has emerged through the concept of complex networks. Networks representing relationships between individual objects usually exhibit community dynamics. Conventional community detection methods mainly focus on either mining frequent subgraphs in a network or detecting stable communities in time-varying networks. In this paper, we formulate a novel problem—detection of predictive and phase-biased communities in contrasting groups of networks, and propose an efficient and effective machine learning solution for finding such anomalous communities. We build different groups of networks corresponding to different system’s phases, such as higher or low hurricane activity, discover phase-related system components as seeds to help bound the search space of community generation in each network, and use the proposed contrast-based technique to identify the changing communities across different groups. The detected anomalous communities are hypothesized (1) to play an important role in defining the target system’s state(s) and (2) to improve the predictive skill of the system’s states when used collectively in the ensemble of predictive models. When tested on the two important extreme event problems—identification of tropical cyclone-related and of African Sahel rainfall-related climate indices—our algorithm demonstrated the superior performance in terms of various skill and robustness metrics, including 8–16 % accuracy increase, as well as physical interpretability of detected communities. The experimental results also show the efficiency of our algorithm on synthetic datasets.     

Fast multi-scale detection of overlapping communities using local criteria

Many systems can be described using graphs, or networks. Detecting communities in these networks can provide information about the underlying structure and functioning of the original systems. Yet this detection is a complex task and a large amount of work was dedicated to it in the past decade. One important feature is that communities can be found at several scales, or levels of resolution, indicating several levels of organisation. Therefore a graph may have several community structures. Also networks tend to be large and hence require efficient processing. In this work, we present a new algorithm with linear complexity for the fast detection of communities across scales using a local criterion. We exploit the local aspect of the criterion to enable parallel computation and improve the execution speed further. The algorithm is tested against very large generated multi-scale networks and experiments demonstrate its efficiency and accuracy.