Algorithm For Community Detection In Networks Research Articles

Carbon Communities and Hotspots for Carbon Emissions Reduction in China

With China’s commitment to peak its emissions by 2030, sectoral emissions are under the spotlight due to the rolling out of the national emission trading scheme (ETS). However, the current sector policies focus either on the production side or consumption while the majority of sectors along the transmission were overlooked. This research combines input–output modelling and network analysis to track the embodied carbon emissions among thirty sectors of thirty provinces in China. Based on the large-data resolution network, a two-step network reduction algorithm is used to extract the backbone of the network. In addition, network centrality metrics and community detection algorithms are used to assess each individual sector’s roles, and to reveal the carbon communities where sectors have intensive emission links. The research results suggest that the sectors with high out-degree, in-degree or betweenness can act as leverage points for carbon emissions mitigation. In addition to the electricity sector, which is included in the national ETS, the study also found that the metallurgy and construction sectors should be prioritized for emissions reduction from national and local levels. However, the hotpots are different across provinces and thus provincial specific targeted policies should be formed. Moreover, there are nineteen carbon communities in China with different features, which provides direction for provincial governments’ external collaboration for synergistic effects.

Research on Community Center-metric and Community Detection Algorithm for Complex Networks

To resolved the present-existing problem that local community detection is great sensitive to the initial node position in large-scale network, a new community detection algorithm CCMA (Community Central Metric Algorithm) by using the community central node-set as the initial node is proposed. Firstly, community centrality node-set (LC ≥ 0) is determined by using Leverage Centrality (LC). Then, starting from the node with the largest central index, we expand the community according to the community fitness function until completing the acquisition of the entire local community. And community’s numbers does not need to be given. Experiments on existing real networks and artificial networks show that our community central metric algorithm can detect local community structures more efficiently and accurately by comparing the method of expanding community’s central node with the existing best local community detection method.

Topological and functional comparison of community detection algorithms in biological networks

BackgroundCommunity detection algorithms are fundamental tools to uncover important features in networks. There are several studies focused on social networks but only a few deal with biological networks. Directly or indirectly, most of the methods maximize modularity, a measure of the density of links within communities as compared to links between communities.ResultsHere we analyze six different community detection algorithms, namely, Combo, Conclude, Fast Greedy, Leading Eigen, Louvain and Spinglass, on two important biological networks to find their communities and evaluate the results in terms of topological and functional features through Kyoto Encyclopedia of Genes and Genomes pathway and Gene Ontology term enrichment analysis. At a high level, the main assessment criteria are 1) appropriate community size (neither too small nor too large), 2) representation within the community of only one or two broad biological functions, 3) most genes from the network belonging to a pathway should also belong to only one or two communities, and 4) performance speed. The first network in this study is a network of Protein-Protein Interactions (PPI) in Saccharomyces cerevisiae (Yeast) with 6532 nodes and 229,696 edges and the second is a network of PPI in Homo sapiens (Human) with 20,644 nodes and 241,008 edges. All six methods perform well, i.e., find reasonably sized and biologically interpretable communities, for the Yeast PPI network but the Conclude method does not find reasonably sized communities for the Human PPI network. Louvain method maximizes modularity by using an agglomerative approach, and is the fastest method for community detection. For the Yeast PPI network, the results of Spinglass method are most similar to the results of Louvain method with regard to the size of communities and core pathways they identify, whereas for the Human PPI network, Combo and Spinglass methods yield the most similar results, with Louvain being the next closest.ConclusionsFor Yeast and Human PPI networks, Louvain method is likely the best method to find communities in terms of detecting known core pathways in a reasonable time.

Brain network analysis for auditory disease: A twofold study

Tinnitus is the phantom perception of sound and can have negative effect on the quality of life. Sudden deafness is generally defined as sensorineural hearing loss of 30 dB or greater over at least three contiguous audiometric frequencies and within a three-day period, which is usually unilateral and can be associated with tinnitus. The exact pathogenesis of such auditory diseases is still unknown, and alterations in the functional connectivity of the brain network are suspected to involve one possible pathogenesis. Scalp electroencephalogram (EEG) analysis is a common approach to capture brain neural activity. This paper conducts a twofold study in terms of brain network analysis for auditory diseases based on EEG data. Specifically, the brain network is constructed in view of the phase lag index (PLI) between signals. The unsupervised study is first conducted via the unsupervised network community detection algorithm. The goal is to reveal the underlying characteristics of functional connectivity concerning subjects with and without auditory diseases from the brain network perspective. The second study is the supervised learning. A new strategy termed FBA (Features from Brain Areas) is proposed to abstract each brain network as a feature vector and the supervised classification method, i.e., Support Vector Machine (SVM) is utilized to perform the three-class classification concerning the regular control group, tinnitus patient group and sudden deafness patient group. Extensive experiments have been conducted on totally 146 subjects including 41 regular controls, 51 sudden deafness patients and 54 tinnitus patients. The results show the necessary of the unsupervised study that may shed light on the pathogenesis and the classification performed in the supervised study may serve as a portable detection method for auditory diseases.

Network model-based analysis of the goals, targets and indicators of sustainable development for strategic environmental assessment

Strategic environmental assessment is a decision support technique that evaluates policies, plans and programs in addition to identifying the most appropriate interventions in different scenarios. This work develops a network-based model to study interlinked ecological, economic, environmental and social problems to highlight the synergies between policies, plans, and programs in environmental strategic planning. Our primary goal is to propose a methodology for the data-driven verification and extension of expert knowledge concerning the interconnectedness of the sustainable development goals and their related targets. A multilayer network model based on the time-series indicators of the World Bank open data over the last 55 years was assembled. The results illustrate that by providing an objective and data-driven view of the correlated variables of the World Bank, the proposed layered multipartite network model highlights the previously not discussed interconnections, node centrality measures evaluate the importance of the targets, and network community detection algorithms reveal their strongly connected groups. The results confirm that the proposed methodology can serve as a data-driven decision support tool for the preparation and monitoring of long-term environmental policies. The developed new data-driven network model enables multi-level analysis of the sustainability (goals, targets, indicators) and will make it possible to plan long-term environmental strategic planning. Through relationships among indicators, relationships among targets and goals can be modelled. The results show that sustainable development goals are strongly interconnected, while the 5th goal (gender equality) is linked mostly to 17th (partnerships for the goals) goal. The analysis has also highlighted the importance of the 4th (quality education).

Community Detection Algorithm for Complex Networks Based on Group Density

Community Detection Algorithm for Complex Networks Based on Group Density

A Scheme to Design Community Detection Algorithms in Various Networks

Network structures, consisting of nodes and edges, have applications in almost all subjects. A set of nodes is called a community if the nodes have strong interrelations. Industries (including cell phone carriers and online social media companies) need community structures to allocate network resources and provide proper and accurate services. However, most detection algorithms are derived independently, which is arduous and even unnecessary. Although recent research shows that a general detection method that serves all purposes does not exist, we believe that there is some general procedure of deriving detection algorithms. In this paper, we represent such a general scheme. We mainly focus on two types of networks: transmission networks and similarity networks. We reduce them to a unified graph model, based on which we propose a method to define and detect community structures. Finally, we also give a demonstration to show how our design scheme works.

Overlapping community detection based on conductance optimization in large-scale networks

Community structure reveals useful information in domains of sociology, biology, physics and computer science. In this work, an overlapping community detection algorithm for large-scale networks based on local expansion is proposed, in which we present a novel seeding method. And we optimize conductance of communities by: (1) modifying inaccurate community affiliations by node movements; (2) combining densely overlapping communities with a novel combining function; (3) finding communities for the outliers with our proposed theorem. Experimental results in synthetic networks show that the optimization largely enhance the community accuracy. Experimental results in large real-world networks show that our approach is superior to the others in the state of the art.

Detection of Communities in a Graph of Interactive Objects

This article describes the problem of analysis of social network graphs and other interacting objects. It also presents community detection algorithms in social networks and their classification and analysis. In addition, it considers applicability of algorithms for real tasks in social network graph analysis.

A complex network community detection algorithm based on label propagation and fuzzy C-means

Community detection algorithms have important significance in the research and practical application of complex network theory. This paper proposes a community detection method by improved label propagation and fuzzy C-means. Due to low accuracy and instability detection results, we modify original label propagation framework. Primarily, initial labels of vertexes are assigned by neighbor evaluation method. Secondarily, the labels of vertexes with large diversity in each community are revised by fuzzy C-means membership vectors. Tertiarily, parameters are updated until communities status is stabilized ultimately. The results showed that this method can achieve better accuracy on synthetic and real network.

Low-rank subspace learning based network community detection

Community structure is one of the most important inherent properties in complex networks. Generally, nodes in each community span a different subspace in the geometric space. However, the subspace based methods are not well studied, and the existing related algorithms are sensitive to the perturbations in networks, thus lack the ability to capture the robust subspace segmentation accurately. In this paper we propose a low-rank subspace learning based network community detection algorithm. The contribution of this paper is to find the lowest rank representation of all node vectors jointly in the geometric space using low-rank decomposition strategy. Then the final robust low-rank subspace based communities are obtained from the lowest rank representations. Low-rank decomposition is better at capturing global structure of networks, thus our algorithm is more robust to perturbations in networks and has a better performance on discriminating the community boundaries. Experimental results on both synthetic benchmarks and real-world networks demonstrate that the proposed algorithm outperforms the state-of-the-art community detection approaches and the results basically preserve the actual community structure, especially on networks with vague community structures.

DASSCAN: A Density and Adjacency Expansion-Based Spatial Structural Community Detection Algorithm for Networks

Existing spatial community detection algorithms are usually modularity based. Motivated by different applications, these algorithms build appropriate spatial null models to describe spatial effects on the connection of nodes. Then, by choosing certain modularity maximizing strategies, they try to find interesting community structures hidden behind the null models. In this paper, a novel structural similarity-based spatial network community is defined, which is based on the shared neighbors of nodes. In addition, there are two other special node roles defined: the spatial hub and outlier. Then, a density and adjacency expansion-based spatial structural community detection algorithm for networks (DASSCAN) is proposed for mining these communities, hubs and outliers. DASSCAN uses structural similarity to measure the relationship between nodes, and then, structurally similar and spatially adjacent nodes are merged into communities using a density-based clustering method and spatial adjacency expansion strategy. Comparative experiments on two kinds of Chinese train line networks clarified the accuracy and efficiency of DASSCAN in finding the spatial structural communities, spatial hubs and outliers. The communities found can be used to uncover more interesting spatial structural patterns, and the hubs and outliers are more accurate and have more valuable meanings.

A Probabilistic Framework for Structural Analysis and Community Detection in Directed Networks

There is growing interest in structural analysis of directed networks. Two major points that need to be addressed are: 1) a formal and precise definition of the graph clustering and community detection problem in directed networks and 2) algorithm design and evaluation of community detection algorithms in directed networks. Motivated by these, we develop a probabilistic framework for structural analysis and community detection in directed networks based on our previous work in undirected networks. By relaxing the assumption from symmetric bivariate distributions in our previous work to bivariate distributions that have the same marginal distributions in this paper, we can still formally define various notions for structural analysis in directed networks, including centrality, relative centrality, community, and modularity. We also extend three commonly used community detection algorithms in undirected networks to directed networks: the hierarchical agglomerative algorithm, the partitional algorithm, and the fast unfolding algorithm. These are made possible by two modularity preserving and sparsity preserving transformations. In conjunction with the probabilistic framework, we show these three algorithms converge in a finite number of steps. In particular, we show that the partitional algorithm is a linear time algorithm for large sparse graphs. Moreover, the outputs of the hierarchical agglomerative algorithm and the fast unfolding algorithm are guaranteed to be communities . These three algorithms can also be extended to general bivariate distributions with some minor modifications. We also conduct various experiments by using two sampling methods in directed networks: 1) PageRank and 2) random walks with self-loops and backward jumps.

Similarity-based local community detection for bipartite networks

Local community detection aims to obtain the local communities to which target nodes belong, by employing only partial information of the network. As a commonly used network model, bipartite applies naturally when modeling relations between two different classes of objects. There are three problems to be solved in local community detection, such as initial core node selection, expansion approach and community boundary criteria. In this work, a similarity based local community detection algorithm for bipartite networks (SLCDB) is proposed, and the algorithm can be used to detect local community structure by only using either type of nodes of a bipartite network. Experiments on real data prove that SLCDB algorithms output community structure can achieve a very high modularity which outperforms most existing local community detection methods for bipartite networks.

Comparing Community Detection Algorithms in Transport Networks via Points of Interest

Passengers travel in transport networks with diverse interests represented by linked points of interest (POIs) and drive urban regions to group into network communities. Previous studies focused on applying community detection methods (CDMs) to discover spatial mobility patterns or using POIs to explain the decision making of human mobility, without comparing the effectiveness of CDMs for detecting network communities. In this paper, we analyze the relationship between POIs and network communities of human mobility over diverse CDMs. Taking the taxi systems of Shanghai and Beijing as case studies, we construct transport networks with urban regions as nodes and the connections between them as links weighted by mobility flows. The spatial communities are identified based on the movement strength among regions. POIs are mapped into nodes in the network and are considered as independent variables for classifying the spatial community categories. Our study suggests that communities detected with two specific CMDs (namely, the Combo algorithm and the Walktrap algorithm) correlate to POIs, and the correlation of the Combo is the best ( $R^{2}=0.3$ for Shanghai and $R^{2}=0.48$ for Beijing). In this regard, this paper can provide valuable insight into understanding the formation of spatial communities and assist in selecting reasonable CDMs.

Dynamic Social Network Community Detection Algorithm Based on Hidden Markov Model

Dynamic Social Network Community Detection Algorithm Based on Hidden Markov Model

Leader‐based community detection algorithm for social networks

Community detection has become a crucial task in social network mining. Detecting communities summarizes interactions between members for gaining deep understanding of interesting characteristics shared between members of the same community. In this research, we propose a novel community detection algorithm for the purpose of revealing and analyzing hidden similar behavior of online users. The proposed algorithm is based mainly on similar members’ actions rather than the structure similarity only for the aim of detecting communities that are closely mapped to the underlying behavioral communities in real social networks. First, leaders of the social network are discovered, then, communities are detected based on those leaders. The idea is grounded on the assumption that communities could be formed around people with great influence. Extensive experiments and analysis show the ability of the proposed algorithm to successfully detect real‐world communities with improved accuracy. WIREs Data Mining Knowl Discov 2017, 7:e1213. doi: 10.1002/widm.1213This article is categorized under: Technologies > Structure Discovery and Clustering

An upper approximation based community detection algorithm for complex networks

The emergence of multifarious complex networks has attracted researchers and practitioners from various disciplines. Discovering cohesive subgroups or communities in complex networks is essential to understand the dynamics of real-world systems. Researchers have made persistent efforts to investigate and infer community patterns in complex networks. However, real-world networks exhibit various characteristics wherein existing communities are not only disjoint but are also overlapping and nested. The existing literature on community detection consists of limited methods to discover co-occurring disjoint, overlapping and nested communities.In this work, we propose a novel rough set based algorithm capable of uncovering true community structure in networks, be it disjoint, overlapping or nested. Initial sets of granules are constructed using neighborhood connectivity around the nodes and represented as rough sets. Subsequently, we iteratively obtain the constrained connectedness upper approximation of these sets. To constrain the sets and merge them during each iteration, we utilize the concept of relative connectedness among the nodes. We illustrate the proposed algorithm on a toy network and evaluate it on fourteen real-world benchmark networks. Experimental results show that the proposed algorithm reveals more accurate communities and significantly outperforms state-of-the-art techniques.

Network analysis and systemic FX settlement risk

Abstract A proposal for applying network analysis to a foreign exchange (FX) settlement system is considered. In particular, network centrality metrics are used to analyse payments of financial institutions which settle through CLS Bank (CLS). Network centrality metrics provide a way to study settlement members’ connectivity, obtain a sense of their payments evolution with time, and measure their network topology variability. The analysis shows that although the continuous link settlement (CLS) network structure can be approximated with a power law degree distribution for many trade days, this is not always the case. A network community detection algorithm is applied to the FX settlement network to explore relationships between communities and to detect classification patterns in the FX trading net payments. A metric called SinkRank is used to build a ranking of the most systemic settlement risk important financial institutions trading on the FX system, and to understand how the metric depends on network’s connectivity. Since network metrics do not fully explain the dynamics of the settlement process, the CLS’ settlement system is simulated to measure the contagion of unsettled trades and its spread among network members. The effect of settlement failure and contagion on the settlement members is also explored.

Adaptive Label Propagation Algorithm to Detect Overlapping Community in Complex Networks

According to the defects that community detection algorithm in unknown complex networks has a pre-parameter. We propose Adaptive Label Propagation Algorithm (ALPA) to detect community structures in complex networks. The ALPA algorithm find out all disjoint Maximal Clique (MC) and let each MC share the identical weight and unique label so as to reduce the redundant labels and uncontrollable factors. The stability of ALPA algorithm is enhanced by synchronous update during iterations. Meanwhile it will converge easily due to the termination condition that all of the vertexes have the label. During iterations we use the adaptive threshold method to overcome the pre-parameter limitation. Compared with other community detection algorithms in synthetic networks and real networks, our experiments show that ALPA algorithm not only improves the tolerance of mixing parameter, but also enhances its robustness.