Communities In Real Networks Research Articles

The application of clique percolation method in health disparities research.

Clique percolation, one of the joint community detection algorithms in network science, is a novel and efficient approach to detecting overlapping communities in real networks. The current study illustrated how clique percolation can help to identify overlapping communities within the complex networks underlying health disparities, particularly highlighting nodes with strong associations with more than one community. A cross-sectional study. The study used a dataset on Latinx populations (N = 1654; mean age = 43.3 years; 53.1% women) as an example to demonstrate the role of such overlapping nodes in the network of syndemic conditions and their common risk factors. Syndemic conditions in the network included HIV risk, substance abuse (smoking, heavy alcohol consumption and marijuana use) and poor mental health. Moreover, the risk factors encompassed individual (education and income) and sociostructural (adverse childhood experiences [ACEs] and access to services) factors. The network was estimated using the R-package bootnet. Clique percolation was conducted on the estimated network using the R-package CliquePercolation. A total of three communities were detected, with HIV risk and poor mental health not being assigned to any community. In general, Community 1 was comprised of ACE categories, Community 2 included education, income and access to services and Community 3 included other syndemic conditions. Of note, two nodes were assigned to two communities: 'household dysfunction' to Communities 1 and 2 and 'smoking' to Communities 2 and 3. Household dysfunction might be the key connector, among other ACEs, to individual and structural barriers. Such barriers further exposed Latinx individuals to risky behaviours, especially smoking, which further linked to marijuana use and heavy alcohol consumption. Clique percolation facilitated our understanding of the complex systems of factors shaping health disparities. The overlapping nodes are promising intervention targets for reducing health disparities in this historically marginalized population. No patient or public contribution.

Community detection using Local Group Assimilation

Clustering of vertices in complex networks to detect communities is an open challenge due to its unknown and hidden properties and broad areas. Complex networks occur in various areas and interdisciplinary fields be it biological, social, chemical, electrical or any other. Numerous methods have been proposed with significant contributions to detect communities but still its nature of properties, throw us a challenge in detecting meaningful communities in real networks. Communities are built through nodes and each node has its own significance. Similarity among nodes through their neighbors suggests common interest of those nodes. The idea is to identify small groups of similar nodes locally and gradually built communities using global information. Moreover, we are not taking any seed nodes and consider each node an important player. The identification of local structures and demarcating their boundaries possess another challenge. Here, we propose Local Group Assimilation (LGA) algorithm that identifies clusters or communities in a network graph using both local and global structure information. The algorithm compares two adjacent nodes by neighborhood similarity measure and picks the highest value pair. The highest value pairs of nodes are grouped together in such a manner that it generates initial clusters of various sizes. The local groups are merged further in iterative manner that maximizes inter cluster edge density between them. Our algorithm detects small significant and relevant communities on real networks that assimilate into larger ones with promising results. We evaluated our algorithm in both real world networks and synthetic benchmark networks and compared it with most popular state-of-the-art community detection algorithms. Our experimental results show that the LGA algorithm detects significant communities in complex networks comparable to most popular algorithms and can be used in real networks to detect communities.

A three-stage algorithm on community detection in social networks

Detecting communities or clusters of networks is a considerable interesting problem in various fields and interdisciplinary subjects in recent years. Tens of hundreds of methods with significant efforts devoted to community detection in networks, while an open problem in all methods is the unknown number of communities in real networks. It is believed that the central node in a community might be highly surrounded by its neighbors and any two centers of the community reside far from each other, and also believed the similarity among nodes in the same community is larger than the others. Therefore, the local and the global structures’ information shed important light on community detection. In this work, we present a three-stage algorithm to detect communities based on the local and the global information without giving the number of communities beforehand. The three stages include the central nodes identification, the label propagation and the communities combination. The central nodes are identified according to the distance between them larger than the average; the label propagation is to label nodes with the same colors when they reach to the maximum similarity; the communities combination is to merge two communities into one if the increment of the modularity is positive and maximum when the two communities were combined. Experiments and simulation results both on real world and synthetic networks show that the three-stage algorithm possesses well matched properties compared with seven other widely used algorithms, which indicates that three-stage algorithm can be used to detect community in social networks.

Multilayer overlapping community detection using multi-objective optimization

Abstract Community detection in multilayer networks such as social or information networks, due to its vast practical applications has attracted many attentions in the past years. Although some researches have been done to develop monoplex methods to multilayers, but because of the complexity of multilayer networks they are in their infancy. In this study, initially, the definition of community in single layer networks is extended and a new definition for multilayer community is presented. Then, regarding the importance of overlapping communities in real networks, a comprehensive definition for overlapping multilayer community is presented. Furthermore, in order to address the problem of multilayer community detection, a two-phase approach has been adopted. In the first phase, a multi-objective mathematical model is developed, and optimized using a genetic algorithm based method (NSGA-II) to achieve a set of solutions (Pareto front). Each potential solution represents a partition of the multilayer network. In the second phase, the best solution is chosen from the Pareto fronts, using a novel algorithm. This proposed algorithm, handles the obstacles of genetic representation and is able to detect the optimal multilayer overlapping communities. Experiments on both synthetic mLFR networks with different parameters and six real data networks show the performance of the algorithm in terms of NMI and multilayer modularity.

Multipolarization versus unification in community networks

Community structure and core–periphery structure are two natural properties of complex networks. Both structures have been studied separately for decades. However, few researchers focus on the combination of these two important structures in complex networks. In this paper, we explore the core–periphery structures of communities in complex networks especially community networks, more precisely, we propose a linear algorithm to divide each community into a densely interconnected core and a periphery where the nodes are rarely linked to each other. Based on core–periphery structures, we perform quantitative analysis of the edges between different communities and find two relationships of two communities in real networks: unitive and multipolar. Communities are called unitive if edges between different cores are more than edges between different peripheries. Otherwise, communities are called multipolar. Furthermore, we propose a random model called Generalized Girvan–Newman(GGN) model, which can generate community networks where communities are either unitive or multipolar. The model sheds some new light on community formation and core–periphery structures in complex systems.

Parameterized Algorithms for the H-Packing with t-Overlap Problem

We introduce the k-H-Packing with t-Overlap problem to formalize the problem of discovering overlapping communities in real networks. More precisely, in the k-H-Packing with t-Overlap problem, we search in a graph G for at least k subgraphs each isomorphic to a graph H such that any pair of subgraphs shares at most t vertices. In contrast with previous work where communities are disjoint, we regulate the overlap through a variable t. Our focus is on the parameterized complexity of the k-H-Packing with t-Overlap problem. Here, we provide a new technique for this problem generalizing the crown decomposition technique [2]. Using our global rule, we achieve a kernel with size bounded by 2(rk r) for the k-Kr-Packing with (r 2)Overlap problem. That is, when H is a clique of size r and t = r 2. In addition, we introduce the first parameterized algorithm for the kH-Packing with t-Overlap problem when H is an arbitrary graph of size r. Our algorithm combines a bounded search tree with a greedy localization technique and runs in time O(r rk k (r t 1)k+2 n r ), where n = |V (G)|, r = |V (H)|, and t < r. Finally, we apply this search tree algorithm to the kernel obtained for the k-Kr-Packing with (r 2)-Overlap problem, and we show that this approach is faster than applying a brute-force algorithm in the kernel. In all our results, r and t are constants.