Overlapping Community Structure Research Articles

A new approach to finding overlapping community structure in signed networks based on Neutrosophic theory

Abstract One of the longstanding challenges in network science is the identification of overlapping community structures. Real-world networks often exhibit a complex interplay of positive and negative relationships, making the recognition of overlapping communities a crucial area of research. Current community detection methods in signed networks primarily focus on discovering disjoint communities, where each node belongs exclusively to a single community. However, these algorithms often fail to detect overlapping communities, where nodes can belong to multiple communities simultaneously. To address this limitation, we propose a novel approach called Neutrosophic c-means Overlapping Community Detection (NOCD) based on neutrosophic set (NS) theory. By incorporating NS theory, our approach effectively handles the uncertainty associated with ambiguous community boundaries and appropriately handles nodes on the community boundaries and isolated nodes. The NOCD method comprises two phases: firstly, a signed graph convolutional neural network is employed to learn the structural features of the signed network in a lower-dimensional representation; secondly, overlapping communities are detected using the neutrosophic c-means algorithm applied to the embedded network. To evaluate the effectiveness of our proposed NOCD method, we conducted comprehensive experiments on both real and artificial networks. The experimental results demonstrate the effectiveness and robustness of NOCD in identifying overlapping communities, outperforming existing methods. [Received on 3 August 2023; editorial decision on 13 December 2023; accepted on 19 December 2023]

FLPA: A fast label propagation algorithm for detecting overlapping community structure

FLPA: A fast label propagation algorithm for detecting overlapping community structure

Overlapping community detection in software ecosystem based on pheromone guided personalized PageRank algorithm

Software ecosystem has aroused the interest of numerous researchers and plays an important role in many aspects. According to different participants and their relationships, software ecosystem can be constructed into various types of complex networks. Overlapping community detection in complex networks can help reveal the community structure and find the intersection between communities. However, existing overlapping community detection algorithms often suffer from reduced applicability or accuracy when applied to networks in software ecosystems. To reveal the overlapping community structure in software ecosystem, we propose an overlapping community detection algorithm by improving the standard personalized PageRank (PPR) algorithm. We first construct a developer collaboration network in software ecosystem based on the intensity of cooperation between developers. Then, the similarity between developers is calculated to guide the walking process of the PPR algorithm, making it suitable for weighted networks. Finally, in the proposed algorithm PGPPR, inspired by the idea of using pheromones to guide the walking in ant colony algorithm, we run the algorithm in multiple rounds and use the results of the previous round to guide the walking process in the current round to reduce redundant diffusion. The experimental results on the five real-world networks show that our algorithm is applicable and effective in detecting communities. And in the five developer collaboration networks, PGPPR can effectively detect overlapping community structures with higher stability and accuracy than the four baselines. The PGPPR algorithm can find overlapping community structures in weighted networks and effectively reduce the redundant diffusion generated when applying the standard PPR algorithm to community detection. Compared to other algorithms, our algorithm can detect the overlapping communities more accurately and stably when applied to developer collaboration networks in software ecosystem.

A one-stage memetic algorithm for jointly detecting hierarchical and overlapping community structures in dynamic social networks

A one-stage memetic algorithm for jointly detecting hierarchical and overlapping community structures in dynamic social networks

The Overlapping Community-Driven Feedback Mechanism to Support Consensus in Social Network Group Decision Making

Social network group decision-making (SN-GDM) provides valuable support for obtaining agreed decision results by effectively utilizing the connected social trust relationships among individuals. However, the impact of intricately overlapped social trust relationships within overlapping communities on evaluation modifications in the SN-GDM consensus reaching process is seldom considered. To alleviate this issue, this study attempts to construct an overlapping community driven feedback mechanism for improving consensus in SN-GDM. The Lancichinetti-Fortunato method (LFM) is used to detect the overlapping community structures under social trust networks. Subsequently, the trusted recommendation advice is conducted within overlapping communities, which guides the inconsistent subgroups to make an interaction with each other to reach higher consensus level. Then, an associated feedback mechanism for SN-GDM with overlapping communities is proposed, which enables the inconsistent subgroups to minimize the consensus cost by selecting personalized feedback parameters. Moreover, it shows that the overlapping communities based feedback mechanism is superior to the feedback mechanisms with non-overlapping communities. Finally, an illustrative example is included, which is also used to testify the efficacy of proposal by comparing the consensus cost under different representative recommendation advice in overlapping social trust networks.

Large-scale community detection based on core node and layer-by-layer label propagation

With the vigorous development of big data, the task of large-scale community structure detection is more challenging. In this paper, a large-scale community detection method based on core node and layer-by-layer label propagation is proposed, which is further extended to the detection of overlapping community structures. First, the core nodes whose node degree is greater than the average degree in the graph are found to make effective use of the feature that the core node is the potential community center. This will also avoid the impact of nodes with low node degrees on community structure detection. Then, starting from the core node, label propagation is carried out layer-by-layer according to the node degree and node connection, which effectively improves the accuracy of community detection. The node labels after label propagation are calibrated according to the current attraction of the community to the nodes, which effectively improves the situation of misclassification in the early community detection. Finally, overlapping community detection is carried out based on the non-overlapping community structure. At this time, the overlapping community detection result is more accurate and interpretable. The community detection results on 3 synthetic networks and 12 real datasets show that the proposed algorithm has more advantages than four non-overlapping community detection methods and two overlapping community detection methods.

A time evolving online social network generation algorithm

The rapid growth of online social media usage in our daily lives has increased the importance of analyzing the dynamics of online social networks. However, the dynamic data of existing online social media platforms are not readily accessible. Hence, there is a necessity to synthesize networks emulating those of online social media for further study. In this work, we propose an epidemiology-inspired and community-based, time-evolving online social network generation algorithm (EpiCNet), to generate a time-evolving sequence of random networks that closely mirror the characteristics of real-world online social networks. Variants of the algorithm can produce both undirected and directed networks to accommodate different user interaction paradigms. EpiCNet utilizes compartmental models inspired by mathematical epidemiology to simulate the flow of individuals into and out of the online social network. It also employs an overlapping community structure to enable more realistic connections between individuals in the network. Furthermore, EpiCNet evolves the community structure and connections in the simulated online social network as a function of time and with an emphasis on the behavior of individuals. EpiCNet is capable of simulating a variety of online social networks by adjusting a set of tunable parameters that specify the individual behavior and the evolution of communities over time. The experimental results show that the network properties of the synthetic time-evolving online social network generated by EpiCNet, such as clustering coefficient, node degree, and diameter, match those of typical real-world online social networks such as Facebook and Twitter.

Overlapping community‐based particle swarm optimization algorithm for influence maximization in social networks

AbstractInfluence maximization, whose aim is to maximise the expected number of influenced nodes by selecting a seed set of k influential nodes from a social network, has many applications such as goods advertising and rumour suppression. Among the existing influence maximization methods, the community‐based ones can achieve a good balance between effectiveness and efficiency. However, this kind of algorithm usually utilise the network community structures by viewing each node as a non‐overlapping node. In fact, many nodes in social networks are overlapping ones, which play more important role in influence spreading. To this end, an overlapping community‐based particle swarm optimization algorithm named OCPSO for influence maximization in social networks, which can make full use of overlapping nodes, non‐overlapping nodes, and their interactive information is proposed. Specifically, an overlapping community detection algorithm is used to obtain the information of overlapping community structures, based on which three novel evolutionary strategies, such as initialisation, mutation, and local search are designed in OCPSO for better finding influential nodes. Experimental results in terms of influence spread and running time on nine real‐world social networks demonstrate that the proposed OCPSO is competitive and promising comparing to several state‐of‐the‐arts (e.g. CGA, CMA‐IM, CIM, CDH‐SHRINK, CNCG, and CFIN).

Data-driven static and dynamic resilience assessment of the global liner shipping network

As a critical infrastructure system of modern society, the global liner shipping network (GLSN) has become increasingly complex and thus vulnerable to disruptions. This study proposes a resilience assessment framework for the GLSN across two dimensions, including static resilience and dynamic resilience. First, by leveraging high-frequency vessel movement data, the GLSN is constructed. Then, with the clique percolation method (CPM), overlapping community structures and key nodes can be identified. The static resilience assessment is initially conducted using simulation techniques, with nodes attacked through pre-designed scenarios. Then, a network disintegration method is employed to consider the impact of traffic flow on system resilience assessment, which separates the weighted GLSN into different layers for evaluation. The results show that both overlapping community structure and traffic flow significantly impact the resilience evaluation of the GLSN. Finally, to assess the dynamic resilience of the GLSN, we propose an innovative, knock-on effect simulation model with tailored, locally weighted flow redistribution rules. It provides a method for predicting the impacts of potential global disruptions (e.g., the COVID-19 pandemic) and critical maritime infrastructure failures (e.g., the Suez Canal obstruction) on the shipping network, which are of great concern not only to academia but also to industry.

Identifying Influential Communities Using IID for a Multilayer Networks

In online social networks (OSN), they generate several specific user activities daily, corresponding to the billions of data points shared. However, although users exhibit significant interest in social media, they are uninterested in the content, discussions, or opinions available on certain sites. Therefore, this study aims to identify influential communities and understand user behavior across networks in the information diffusion process. Social media platforms, such as Facebook and Twitter, extract data to analyze the information diffusion process, based on which they cascade information among the individuals in the network. Therefore, this study proposes an influential information diffusion model that identifies influential communities across these two social media sites. Moreover, it addresses site migration by visualizing a set of overlapping communities using hyper-edge detection. Thus, the overlapping community structure is used to identify similar communities with identical user interests. Furthermore, the community structure helps in determining the node activation and user influence from the information cascade model. Finally, the Fraction of Intra/Inter-Layer (FIL) diffusion score is used to evaluate the efficiency of the influential information diffusion model by analyzing the trending influential communities in a multilayer network. However, from the experimental result, it observes that the FIL diffusion score for the proposed method achieves better results in terms of accuracy, precision, recall and efficiency of community detection than the existing methods.

A Hadoop Based Approach for Community Detection on Social Networks Using Leader Nodes

Community detection is the most common and growing area of interest in social and real-time network applications. In recent years, several community detection methods have been developed. Particularly, community detection in Local expansion methods have been proved as effective and efficiently. However, there are some fundamental issues to uncover the overlapping communities. The maximum methods are sensitive to enable the seeds initialization and construct the parameters, while others are insufficient to establish the pervasive overlaps. In this paper, we proposed the new unsupervised Map Reduce based local expansion method for uncovering overlapping communities depends seed nodes. The goal of the proposed method is to locate the leader nodes (seed nodes) of communities with the basic graph measures such as degree, betweenness and closeness centralities and then derive the communities based on the leader nodes. We proposed Map-Reduce based Fuzzy C- Means Clustering Algorithm to derive the overlapping communities based on leader nodes. We tested our proposed method Leader based Community Detection (LBCD) on the real-world data sets of totals of 11 and the experimental results shows the more effective and optimistic in terms of network graph enabled overlapping community structures evaluation.

A Hadoop Based Approach for Community Detection on Social Networks Using Leader Nodes

Community detection is the most common and growing area of interest in social and real-time network applications. In recent years, several community detection methods have been developed. Particularly, community detection in Local expansion methods have been proved as effective and efficiently. However, there are some fundamental issues to uncover the overlapping communities. The maximum methods are sensitive to enable the seeds initialization and construct the parameters, while others are insufficient to establish the pervasive overlaps. In this paper, we proposed the new unsupervised Map Reduce based local expansion method for uncovering overlapping communities depends seed nodes. The goal of the proposed method is to locate the leader nodes (seed nodes) of communities with the basic graph measures such as degree, betweenness and closeness centralities and then derive the communities based on the leader nodes. We proposed Map-Reduce based Fuzzy C- Means Clustering Algorithm to derive the overlapping communities based on leader nodes. We tested our proposed method Leader based Community Detection (LBCD) on the real-world data sets of totals of 11 and the experimental results shows the more effective and optimistic in terms of network graph enabled overlapping community structures evaluation.

A Temporal and Social Network-based Recommender using Graph Clustering

Recommendation Systems (RSs) have significant applications in many industrial systems. The duty of a recommender algorithm is to operate available data (users/items contextual data and rating (or purchase) the consumption history for items), as well as to provide a recommendation list for any target user. The recommended items should be selected so that the target user is compelled to give them positive reviews. In this manuscript, we propose a novel of RS algorithm that makes advantage of user-user trust relationships, rating histories, and their frequency of occurrence. We also provide a brand new overlapping community detection algorithm. The information about the users’ community structure is used to handle the cold-start and sparsity problems. We compare the performance of the proposed RS algorithm with a number of state-of-the-art algorithms on the extended Epinions dataset, which has both information on trust relations and the timing of the ratings. Numerical simulations reveal the superiority of the proposed algorithm over others. We also investigate how the algorithms perform when only cold-start users and items are considered. As a cold-start user (item) we consider those that have made (received) less than five ratings. The experiments show significant outperformance of the proposed algorithm over others, which is mainly due to the use of information on overlapping community structures between users.

Spatio-Temporal Variations of Zooplankton and Correlations with Environmental Parameters around Tiaowei Island, Fujian, China.

The present study illustrates zooplankton dynamics in relation to environmental factors from the surrounding area of Tiaowei Island based on ten seasonal sampling cruises over three years. A total of 116 species of zooplankton were collected with a predominance of Copepoda (mainly consisting of Centropagidae, Oithonidae, Acartia, Labidocera and Paracalanus), accounting for 31.6 % of the total number of species. The diversity indices indicated a relatively high richness, abundance and evenness of zooplankton ranging from 2.794 to 4.012 on the Shannon–Wiener index for each cruise. More than 20 species of Cnidaria medusae are found as gelatinous organisms, which not only compete with fish but also potentially cause disasters. Significant seasonal variations were detected in both the zooplankton structure and environmental variables. NMDS illustrated a highly overlapping community structure in spring, autumn and winter, while the zooplankton composition in the summer was different from that of the other three seasons with a higher diversity index. Meanwhile, out of thirteen environmental parameters, eight varied significantly among seasons but there were no significant variations among stations. The biota–environmental relationship following a redundancy analysis revealed that water temperature, pH, salinity, dissolved oxygen and suspended particulate composition were the main environmental parameters, seasonally impacting the zooplankton communities. Planktonic larvae (such as nauplius larvae and branchyura zoea) and some zooplankton (including Corophium sinensis and Oithona similis) were significantly vulnerable to the dynamics of suspended particulate composition and water temperature.

NeSiFC: Neighbors' Similarity-Based Fuzzy Community Detection Using Modified Local Random Walk.

This article proposes a neighbors' similarity-based fuzzy community detection (FCD) method, which we call "NeSiFC." In the proposed NeSiFC approach, we compute the similarity between two neighbors by introducing a modified local random walk (mLRW). Basically, in a network, a node and its' neighbors with noticeable similarities among them construct a community. To measure this similarity, we introduce a new metric, called the peripheral similarity index (PSI). This PSI is used to construct the transition probability matrix for the mLRW. The mLRW is applied for each node until it meets a parameter called step coefficient. The mLRW gives better neighbors' similarity for community detection. Finally, a fuzzy membership function is used iteratively to compute the membership degrees for all nodes with reference to existing communities. The proposed NeSiFC has no dependence on the network characteristics, and no adjustment or fine tuning of more than one parameter is needed. To show the efficacy of the proposed NeSiFC approach, we provide a thorough comparative performance analysis considering a set of well-known FCD algorithms viz., the genetic algorithm for fuzzy community detection, membership degree propagation, center-based fuzzy graph clustering, FMM/H2, and FuzAg on a set of popular benchmarks, as well as real-world datasets. For both disjoint and overlapping community structures, results of various accuracy and quality metrics indicate the outstanding performance of our proposed NeSiFC approach. The asymptotic complexity of the proposed NeSiFC is found as O(n2).

Research on Overlapping Communities in Scientific Cooperation Network

Science of science provides quantitative means for scientific research activities and can reveal the basic laws of scientific research. Taking the scientist cooperation network of Condensed matter physics, General relativity quantum cosmology and High energy physics phenomenology as the research object, the overlapping community structure in the network is detected by using the label propagation algorithm based on seed diffusion. Through community analysis, it is found that when the scale of scientists' cooperative network increases, the increase in the number of communities is not obvious, but the scale of each community will increase, and the probability of scientists belonging to overlapping communities will decrease. This shows that the increase in the number of authoritative scientists is not obvious after the unpopular research field becomes a popular research field. In hot research fields, the division of scientists' cooperation circles is clearer, and the cooperation frequency between circles will decrease significantly.

An interaction-based method for detecting overlapping community structure in real-world networks

An interaction-based method for detecting overlapping community structure in real-world networks

A Multi-Label Propagation Algorithm With the Double-Layer Filtering Strategy for Overlapping Community Detection

Overlapping community structures are universal in complex networks. Multi-label propagation approach, as a method for detecting overlapping community, has been received widespread attention for the near linear time complexity. In recent years, some algorithms based on multi-label propagation approach have been proposed, but there is still room to improve the accuracy of these algorithms to detect overlapping communities. In this paper, a double-layer filtering strategy is proposed to help nodes filter labels, and the node centrality of each node according to the number of its neighbors and the number of edges between its neighbors is used to fix the node update sequence. Then based on the double-layer filtering strategy, the new fixed node update sequence and the previous multi-label propagation algorithm, a new algorithm called MLPA-DF is proposed. Theory analysis shows that MLPA-DF still has the near linear time complexity. And experimental results on the synthetic and real-world networks show that MLPA-DF has better performance than previous multi-label propagation algorithms (COPRA, SLPA and LPANNI) in terms of the accuracy of detecting overlapping communities without reducing stability.

Biomarkers Derived from Alterations in Overlapping Community Structure of Resting-state Brain Functional Networks for Detecting Alzheimer’s Disease

Recent studies show that overlapping community structure is an important feature of the brain functional network. However, alterations in such overlapping community structure in Alzheimer’s disease (AD) patients have not been examined yet. In this study, we investigate the overlapping community structure in AD by using resting-state functional magnetic resonance imaging (rs-fMRI) data. The collective sparse symmetric non-negative matrix factorization (cssNMF) is adopted to detect the overlapping community structure. Experimental results on 28 AD patients and 32 normal controls (NCs) from the ADNI2 dataset show that the two groups have remarkable differences in terms of the optimal number of communities, the hierarchy of communities detected at different scales, network functional segregation, and nodal functional diversity. In particular, the frontal-parietal and basal ganglia networks exhibit significant differences between the two groups. A machine learning framework proposed in this paper for AD detection achieved an accuracy of 76.7% when using the detected community strengths of the frontal-parietal and basal ganglia networks only as input features. These findings provide novel insights into the understanding of pathological changes in the brain functional network organization of AD and show the potential of the community structure-related features for AD detection.

Deep Learning Exploration of Agent-Based Social Network Model Parameters.

Interactions between humans give rise to complex social networks that are characterized by heterogeneous degree distribution, weight-topology relation, overlapping community structure, and dynamics of links. Understanding these characteristics of social networks is the primary goal of their research as they constitute scaffolds for various emergent social phenomena from disease spreading to political movements. An appropriate tool for studying them is agent-based modeling, in which nodes, representing individuals, make decisions about creating and deleting links, thus yielding various macroscopic behavioral patterns. Here we focus on studying a generalization of the weighted social network model, being one of the most fundamental agent-based models for describing the formation of social ties and social networks. This generalized weighted social network (GWSN) model incorporates triadic closure, homophilic interactions, and various link termination mechanisms, which have been studied separately in the previous works. Accordingly, the GWSN model has an increased number of input parameters and the model behavior gets excessively complex, making it challenging to clarify the model behavior. We have executed massive simulations with a supercomputer and used the results as the training data for deep neural networks to conduct regression analysis for predicting the properties of the generated networks from the input parameters. The obtained regression model was also used for global sensitivity analysis to identify which parameters are influential or insignificant. We believe that this methodology is applicable for a large class of complex network models, thus opening the way for more realistic quantitative agent-based modeling.