Deep Multi-Graph Embedded Clustering for Community Detection in FMRI Functional Brain Networks Across Individuals

Community structures are ubiquitous in various complex networks, implying that the networks commonly be composed of groups of nodes with more internal links and less external links. As an important topic in network theory, community detection is of importance for understanding the structure and function of the networks. Optimizing statistical measures for community structures is one of most popular strategies for community detection in complex networks. In the paper, by using a type of self-loop rescaling strategy, we introduced a set of global modularity functions and a set of local modularity functions for community detection in networks, which are optimized by a kind of the self-consistent method. We carefully compared and analyzed the behaviors of the modularity-based methods in community detection, and confirmed the superiority of the local modularity for detecting community structures on large-size and heterogeneous networks. The local modularity can more quickly eliminate the first-type limit of modularity, and can eliminate or alleviate the second-type limit of modularity in networks, because of the use of the local information in networks. Moreover, we tested the methods in real networks. Finally, we expect the research can provide useful insight into the problem of community detection in complex networks.

Community detection in networks receives much attention recently. Most of the previous works are for unipartite networks composed of only one type of nodes. In real world situations, however, there are many bipartite networks composed of two types of nodes. In this paper, we propose a fast algorithm called LP&BRIM for community detection in large-scale bipartite networks. It is based on a joint strategy of two developed algorithms -- label propagation (LP), a very fast community detection algorithm, and BRIM, an algorithm for generating better community structure by recursively inducing divisions between the two types of nodes in bipartite networks. Through experiments, we demonstrate that this new algorithm successfully finds meaningful community structures in large-scale bipartite networks in reasonable time limit.

With the rapid increase in popularity of online social networks, community detection in these networks has become a key aspect of research field. Overlapping community detection is an important NP-hard problem of social network analysis. Modularity-based community detection is one of the most widely used approaches for social network analysis. However, modularity-based community detection technique may fail to resolve small-size communities. Hence, we propose a novel two-step approach for overlapping community detection in social networks. In the first step, modularity density-based hybrid meta-heuristics approach is used to find the disjoint communities and the quality of these disjoint communities can be verified using Silhouette coefficient. In the second step, the quality disjoint communities with low computation cost are used to detect overlapping nodes based on Min-Max Ratio of minimum(indegree, outdegree) to the maximum(indegree, outdegree) values of nodes. We tested the proposed algorithm based on 10 standard community quality metrics along with Silhouette score using seven standard datasets. Experimental results demonstrate that the proposed approach outperforms the current state-of-the-art works in terms of quality and scalability.

Social network analysis can be an important help for military and criminal intelligence analysis. In real world applications, there is seldom complete knowledge about the network of interest -- we only have partial and incomplete information about the nodes and networks present. Community detection in networks is an important area of current research in social network analysis with many applications. Finding community structures is however a challenging task and despite significant effort no satisfactory method has been found. Here we study the problem of community detection in noisy and uncertain networks with missing and false edges and propose methods for detecting community structures in them. The method is based on sampling from an ensemble of certain networks that are consistent with the available information about the uncertain networks.

With the rapidly growing evidence that various systems in nature and society can be modeled as complex networks, community detection in networks becomes a hot research topic in physics, sociology, computer society, etc. Although this investigation of community structures has motivated many diverse algorithms, most of them are unsuitable when dealing with large networks due to their computational cost. In this paper, we present a faster algorithm ComTector, which is more efficient for the community detection in large complex networks based on the nature of overlapping cliques. This algorithm does not require any priori knowledge about the number or the original division of the communities. With respect to practical applications, ComTector is challenging with five different types of networks including the classic Zachary Karate Club, Scientific Collaboration Network, South Florida Free Word Association Network, Urban Traffic Network, North America Power Grid and the Telecommunication Call Network. Experimental results show that our algorithm can discover meaningful communities that meet both the objective basis and our intuitions.

There is an increasing recognition on community detection in complex networks in recent years. In this study, we improve a recently proposed memetic algorithm for community detection in networks. By introducing a Population Generation via Label Propagation (PGLP) tactic, an Elitism Strategy (ES) and an Improved Simulated Annealing Combined Local Search (ISACLS) strategy, the improved memetic algorithm called (iMeme-Net) is put forward for solving community detection problems. Experiments on both computer-generated and real-world networks show the effectiveness and the multi-resolution ability of the proposed method.

With the advent of the internet and the boom of information sharing, study of networks has begun to play a central role in social science, mathematics and statistical analysis. A large variety of data can be modeled via complex networks making them increasingly more important for research. However, it is not easy to extract meaningful information from a mesh of interconnected of nodes.. That is why, detection of communities in network has become of utmost importance in recent times. Communities can be said to act as meta-nods. They correspond to functional units of the system and hence, often shed light on the function of the system represented by the network. Detecting an underlying community structure in a network thus allows us to create a map of a network which makes it easier to study. In this paper, we examine a number of research involving detection of communities and summarise them based on their avenues of approach to solving the problem.

Introduction Most of neurological disorders are network-based diseases. The networks associated with these diseases usually involve spatially disturbed brain regions. Thus efforts were recently evolving from identifying pathological “zones” toward identifying “networks”. In a very recent review, Fornito and colleagues revealed that the identification of alterations in brain networks is one of the most promising paradigms in brain disorders research (Fornito and Bullmore, 2014; Fornito et al., 2015). So far, approaches based on graph theory have characterized the brain networks as sets of nodes connected by edges (Bullmore and Sporns, 2009). Once the nodes (brain regions) and edges (functional/structural connections between regions) are defined from neuroimaging technique, methods based on graph theory may be used to describe the topological properties of the identified networks. This network-based analysis has been largely used to investigate normal (Bressler and Menon, 2010) and pathological (Fornito et ...

The human brain can be described as a complex network of functional connections between distinct regions, referred to as the brain functional network. Recent studies show that the functional network is a dynamic process and its community structure evolves with time during continuous task performance. Consequently, it is important for the understanding of the human brain to develop dynamic community detection techniques for such time-varying functional networks. Here, we propose a temporal clustering framework based on a set of network generative models and surprisingly it can be linked to Block Component Analysis to detect and track the latent community structure in dynamic functional networks. Specifically, the temporal dynamic networks are represented within a unified three-way tensor framework for simultaneously capturing multiple types of relationships between a set of entities. The multi-linear rank- (Lr, Lr, 1) block term decomposition (BTD) is adopted to fit the network generative model to directly recover underlying community structures with the specific evolution of time from the temporal networks. We apply the proposed method to the study of the reorganization of the dynamic brain networks from electroencephalography (EEG) data recorded during free music listening. We derive several network structures ( Lr communities in each component) with specific temporal patterns (described by BTD components) significantly modulated by musical features, involving subnetworks of frontoparietal, default mode, and sensory-motor networks. The results show that the brain functional network structures are dynamically reorganized and the derived community structures are temporally modulated by the music features. The proposed generative modeling approach can be an effective tool for describing community structures in brain networks that go beyond static methods and detecting the dynamic reconfiguration of modular connectivity elicited by continuously naturalistic tasks.

Human brain is structurally and functionally asymmetrical and the asymmetries of brain phenotypes have been shown to change in normal aging. Recent advances in graph theoretical analysis have showed topological lateralization between hemispheric networks in the human brain throughout the lifespan. Nevertheless, apparent discrepancies of hemispheric asymmetry were reported between the structural and functional brain networks, indicating the potentially complex asymmetry patterns between structural and functional networks in aging population. In this study, using multimodal neuroimaging (resting-state fMRI and structural diffusion tensor imaging), we investigated the characteristics of hemispheric network topology in 76 (male/female = 15/61, age = 70.08 ± 5.30 years) community-dwelling older adults. Hemispheric functional and structural brain networks were obtained for each participant. Graph theoretical approaches were then employed to estimate the hemispheric topological properties. We found that the optimal small-world properties were preserved in both structural and functional hemispheric networks in older adults. Moreover, a leftward asymmetry in both global and local levels were observed in structural brain networks in comparison with a symmetric pattern in functional brain network, suggesting a dissociable process of hemispheric asymmetry between structural and functional connectome in healthy older adults. Finally, the scores of hemispheric asymmetry in both structural and functional networks were associated with behavioral performance in various cognitive domains. Taken together, these findings provide new insights into the lateralized nature of multimodal brain connectivity, highlight the potentially complex relationship between structural and functional brain network alterations, and augment our understanding of asymmetric structural and functional specializations in normal aging.

Previous studies on network mining have focused primarily on learning a single task (such as classification or community detection) on a given network. This paper considers the problem of multi-task learning on heterogeneous network data. Specifically, we present a novel framework that enables one to perform classification on one network and community detection in another related network. Multi-task learning is accomplished by introducing a joint objective function that must be optimized to ensure the classes in one network are consistent with the link structure, nodal attributes, as well as the communities detected in another network. We provide both theoretical and empirical analysis of the framework. We also show that the framework can be extended to incorporate prior information about the correspondences between the clusters and classes in different networks. Experiments performed on both real-world and synthetic data sets demonstrate the effectiveness of the joint framework compared to applying classification and community detection algorithms on each network separately.

Community detection is the identification of different communities or groups that exist within a network. This is useful in social network analysis (SNA) or what is great is performing whole network analysis (WNA), where humans interact with others as part of their various communities, but these approaches are not limited to the study of humans. These methods are to investigate any type of node that interacts closely with other nodes, whether those nodes are animals, hashtags, websites, or any other type of node in the network. In this work, we zoom in on communities that exist in a network. Community detection is a clear, concise, and appropriate name for what we are doing. Communities in the network would be worth exploring and understanding for further purposes. There are several methods and different approaches to detect community, but in this paper, I use two efficient methods to detect whole network which are named Louvain Method (LM) and Girvan-Newman Method (GNM). With LM, we can build a fast algorithm that is effective at community detection in massive networks and optimize the algorithm for better results. Using the GNM, a better approach that can identify the least number of edges that could be cut would result in a split network. We could do this by making an algorithm looking for the edges that the greatest number of shortest paths pass through.

An important task in unsupervised learning is the detection of communities in networks. Although many community detection techniques have been proposed, there are still some challenge problems, such as unbalanced community detection and the low efficiency. In this paper, we propose a community detection technique combining the sequential signal propagation of the Particle Competition model and the parallel propagation inspired by Self-Orgnizing Map (SOM). As a result, the model presents two salient features: 1) It can detect unbalanced communities. 2) It is much more efficient than the original particle competition model due to the introduction of parallel propagation. Still in this work, we analyze functional brain network by identifying the modules (communities) using the proposed technique. Our results show that there is a strong correlation between brain functions and brain regions and a big decrease of intra-strength measure among communities from the Control Network to the Schizophrenia Network, indicating that the functional correlation of brain regions is weakened in the disease network.

Community detection in complex networks plays an important role in mining and analyzing the structure and function of networks. However, traditional algorithms for community detection-based graph partition and hierarchical clustering usually have to face expensive computational costs or require some specific conditions when dealing with complex networks. Recently, community detection based on intelligent optimization attracts more and more attention because of its good effectiveness. In this paper, a new multi-objective ant colony optimization with decomposition (MACOD) for community detection in complex networks is proposed. Firstly, a new framework of multi-objective ant colony algorithm specialized initially for the complex network clustering is developed, in which two-objective optimization problem can be decomposed into a series of subproblems and each ant is responsible for one single objective subproblem and it targets a particular point in the Pareto front. Secondly, a problem-specific individual encoding strategy based on graph is proposed. Moreover, a new efficient local search mechanism is designed in order to improve the stability of the algorithm. The proposed MACOD has been compared with four other state of the art algorithms on two benchmark networks and seven real-world networks including three large-scale networks. Experimental results show that MACOD performs competitively for the community detection problems.

Weighted modularity optimization for crisp and fuzzy community detection in large-scale networks