Community Discovery Methods Research Articles

UCAD: commUnity disCovery method in Attribute-based multicoloreD networks

UCAD: commUnity disCovery method in Attribute-based multicoloreD networks

NDI-PPO: Node degree increment-based proximal policy optimization method for employment community discovery in online social networks

Purpose: aiming at the limitations of pre-input parameters in the complex network overlapping community discovery algorithm based on tag propagation in real networks and the problems of tag redundancy, method: a node degree increment-based proximal policy optimization method for community discovery in online social networks is proposed (named NDI-PPO). Process: by applying the cohesion idea and introducing the concept of modularity increment, a social network great community is constructed from the bottom up according to the criteria of community division. For the problem that the number of iterative steps is sensitive to the strategy gradient algorithm, we adopt an improved PPO to improve the efficiency of feature extraction. In label updating, the maximum clique is used as the core unit to update the labels and weights of the maximum maximum clique adjacent nodes from the center to the periphery using intimacy, and the weights of the non-maximum maximum clique adjacent nodes are updated by means of the maximum weight. In the post-processing stage, the adaptive threshold method is used to remove the noise in the node label, which effectively overcomes the limitation of the number of pre-input overlapping communities in the real network. Result: The simulation results show that the proposed community discovery algorithm NDI-PPO is superior to other advanced algorithms, the time complexity is greatly reduced, and it is suitable for community discovery in large social networks.

A community detection model using node embedding approach and graph convolutional network with clustering technique

The communities in a network have distinct characteristics and interrelationships. Community discovery methods based on node embedding and deep learning have surpassed spectral clustering and statistical inference as the preferred methods for handling high-dimensional network data. Community detection in graph networks using graph neural networks (GNNs) is a rapidly expanding field of study that has attracted much attention recently. Inspired by the graph convolutional network (GCN) approach, promising results in evaluating graph structure data, a different method for discovering communities using node embedding using Node2Vec via graph convolutional networks (CD2NE-GCN) model has been proposed. It has been evaluated against the most recent standard methods of graph attention network (GAT), GCN-based approach for Unsupervised Community Detection (GUCD), and Graph Adversarial Networks (GAN). In this paper, a technique for community detection that combines the ideas of graph neural network design with node embedding via message forwarding has been presented. The experimentation has made use of five benchmark datasets: Karate, Cora, CiteSeer, Facebook, and DBLP. In the first set of trials, the node2vec uses the skip-gram with negative sampling (SGNS) approach via biased random walks of the second order to produce node embeddings and feed them to the graph convolutional network. The weighted average matrix and scaled normalized matrices of low-dimensional vertices have been calculated. In the next set of experiments, the average weighted matrix was sent to the 3-layer GCN with the activation function for generating feature vectors and applied clustering technique for predicting communities. The CD2NE-GCN approach outperforms previous literature by detecting graph network communities with accuracy scores of 98.45%, 99.04%, 96.39%, 98.79%, and 97.67% for Karate, Cora, CiteSeer, Facebook, and DBLP datasets, respectively. Thus, graph convolutional network models with a node embedding approach enhance the accuracy.

Mutual Supervised Fusion & Transfer Learning with Interpretable Linguistic Meaning for Social Data Analytics

Social data analytics is often taken as the most commonly used method for community discovery, product recommendations, knowledge graph, and so on. In this study, social data are firstly represented in different feature spaces by using various feature extraction algorithms. Then we build a transfer learning model to leverage knowledge from multiple feature spaces. During modeling, since the assumption that the training and the testing data have the same distribution is always true, we give a theorem and its proof which asserts the necessary and sufficient condition for achieving a minimum testing error. We also theoretically demonstrate that maximizing the classification error consistency across different feature spaces can improve the classification performance. Additionally, the cluster assumption derived from semi-supervised learning is introduced to enhance knowledge transfer. Finally, aTagaki-Sugeno-Kang (TSK)fuzzy system-based learning algorithm is proposed, which can generate interpretable fuzzy rules. Experimental results not only demonstrate the promising social data classification performance of our proposed approach but also show its interpretability which is missing in many other models.

Relationship Discovery and Hierarchical Embedding for Web Service Quality Prediction.

Web Services Quality Prediction has become a popular research theme in Cloud Computing and the Internet of Things. Graph Convolutional Network (GCN)-based methods are more efficient by aggregating feature information from the local graph neighborhood. Despite the fact that these prior works have demonstrated better prediction performance, they are still challenged as follows: (1) first, the user-service bipartite graph is essentially a heterogeneous graph that contains four kinds of relationships. Previous GCN-based models have only focused on using some of these relationships. Therefore, how to fully mine and use the above relationships is critical to improving the prediction accuracy. (2) After the embedding is obtained from the GCNs, the commonly used similarity calculation methods for downstream prediction need to traverse the data one by one, which is time-consuming. To address these challenges, this work proposes a novel relationship discovery and hierarchical embedding method based on GCNs (named as RDHE), which designs a dual mechanism to represent services and users, respectively, designs a new community discovery method and a fast similarity calculation process, which can fully mine and utilize the relationships in the graph. The results of the experiment on the real data set show that this method greatly improved the accuracy of the web service quality prediction.

Intentional controlled islanding based on dynamic community detection for power grid

Power system controlled islanding is an emergency control method used to stop the propagation of disturbances and avoid blackouts. Intentional controlled islanding (ICI) has been proposed as a corrective measure of last resort to split the power system into several sustainable islands. Complex network community detection is widely used in network partitioning and can be used to solve controlled islanding problems. The objective of this study is to apply the dynamic network community detection method to solve ICI problems. Using dynamic network modelling, the division-agglomeration (Di-Ag) algorithm is proposed. Dynamic modelling presets the computation task of the algorithm in the healthy grid phase. The algorithm balances the efficiency of the algorithm, power imbalance, and power flow disruption, without sacrificing other objectives to optimise one objective. Specifically, the grid topology is taken into account when the grid is divided using betweenness as edge weight. The first step of the Di-Ag algorithm uses the dynamic Girvan–Newman algorithm to achieve the goal of power imbalance. Moreover, only the islands' involved lines (nodes) are updated, which improves the efficiency of the algorithm. In the second step of the Di-Ag algorithm, the algorithm merges the islands to reduce power flow disruption. The IEEE 39-bus and 118-bus system are used to compare the performance of the proposed dynamic modelling method and the static network community discovery method in three different cases.

Advances in scaling community discovery methods for signed graph networks

AbstractCommunity detection is a common task in social network analysis with applications in a variety of fields including medicine, criminology and business. Despite the popularity of community detection, there is no clear consensus on the most effective methodology for signed networks. In this article, we summarize the development of community detection in signed networks and evaluate current state-of-the-art techniques on several real-world datasets. First, we give a comprehensive background of community detection in signed graphs. Next, we compare various adaptations of the Laplacian matrix in recovering ground-truth community labels via spectral clustering in small signed graph datasets. Then, we evaluate the scalability of leading algorithms on small, large, dense and sparse real-world signed graph networks. We conclude with a discussion of our novel findings and recommendations for extensions and improvements in state-of-the-art techniques for signed graph community discovery in real-world signed graphs.

Dynamic Community Discovery Method Based on Phylogenetic Planted Partition in Temporal Networks

As most of the community discovery methods are researched by static thought, some community discovery algorithms cannot represent the whole dynamic network change process efficiently. This paper proposes a novel dynamic community discovery method (Phylogenetic Planted Partition Model, PPPM) for phylogenetic evolution. Firstly, the time dimension is introduced into the typical migration partition model, and all states are treated as variables, and the observation equation is constructed. Secondly, this paper takes the observation equation of the whole dynamic social network as the constraint between variables and the error function. Then, the quadratic form of the error function is minimized. Thirdly, the Levenberg–Marquardt (L–M) method is used to calculate the gradient of the error function, and the iteration is carried out. Finally, simulation experiments are carried out under the experimental environment of artificial networks and real networks. The experimental results show that: compared with FaceNet, SBM + MLE, CLBM, and PisCES, the proposed PPPM model improves accuracy by 5% and 3%, respectively. It is proven that the proposed PPPM method is robust, reasonable, and effective. This method can also be applied to the general social networking community discovery field.

DeMal: Module decomposition of malware based on community discovery

In recent years, malware has grown faster than ever in volume, form and harmfulness. While existing static or dynamic analysis techniques can meet the common user needs for malware detection, analysts desire a more detailed overview to uncover the program architecture. Malware often force research into difficulties due to their complex anti-analysis techniques, which call for a quick analysis of program structure and components to clarify malware functional semantics. In this paper, we use community discovery methods to automate the malware program components analysis from the intuition of modular programing principles. Specifically, we design and implement DeMal, a solution to the malware module decomposition problem. It achieves remodularization by recovering program call relationships, extracting structure-related attributes, and applying an ensemble model of multiple community discovery algorithms. DeMal takes a malicious executable as input and predicts its code composition structure. In an evaluation with 155 malware samples, DeMal performs well on achieving an average F1-score of 71.3%, and 14.5% of the samples reach an average precision of 90%. The analysis time on each sample is about 19.79s. On extended experiments with 1,621 benign programs and over 10,000 stripped malware samples, we also verify DeMal's scalability on common programs as well as the large-scale performance, respectively. The visualization of the results also strongly demonstrates DeMal's module decomposition capabilities.

Study on Power Grid Partition and Attack Strategies Based on Complex Networks

Based on the community discovery method in complex network theory, a power grid partition method considering generator nodes and network weightings is proposed. Firstly, the weighted network model of a power system is established, an improved Fast-Newman hierarchical algorithm and a weighted modular Q function index are introduced, and the partitioning algorithm process is practically improved combined with the characteristics of the actual power grid. Then, the partition results of several IEEE test systems with the improved algorithm and with the Fast-Newman algorithm are compared to demonstrate its effectiveness and correctness. Subsequently, on the basis of subnet partition, two kinds of network attack strategies are proposed. One is attacking the maximum degree node of each subnet, and the other is attacking the maximum betweenness node of each subnet. Meanwhile, considering the two traditional intentional attack strategies, that is, attacking the maximum degree nodes or attacking the maximum betweenness nodes of the whole network, the cascading fault survivability of different types of networks under four attack strategies is simulated and analyzed. It was found that the proposed two attack strategies based on subnet partition are better than the two traditional intentional attack strategies.

Hierarchical community discovery for multi-stage IP bearer network upgradation

The huge and ever-increasing data traffic promotes the upgradation of IP bearer network to possess higher transmission capability. As a long-term task, network upgradation needs to be implemented in multiple stages. At each stage, one or more network partitions are planned to meet various requirements in different areas or periods to realize the smooth change from the current network to the targeted one. However, how to select proper network partitions is an open question. Existing approaches are usually infeasibly applied to large-scale networks or neglect the characteristics of IP bearer network. In this paper, we present a hierarchical community discovery method to find network partitions for nodes to be upgraded. We transform the issue of hierarchy identification on the whole network into node role classification, which is addressed by training a classifier utilizing network centrality, label, and attribute information. Then, a novel community detection algorithm through semi-local expansion is proposed, with the geographical location also taken into consideration. Experiments are conducted on three real-world datasets. The results show the effectiveness of our method in node role classification task that at least 90% Micro-F1 score can be got with only 30% labeled nodes on the given networks. The attained community is adjustable with the size proportional to the parameter α and has a high modularity score. Moreover, our communities strictly comply with the affiliation in IP bearer network and preserve the integrity of the tree/ring structures.

Survey of community discovery method of heterogeneous network

Survey of community discovery method of heterogeneous network

Discovering Community Interests Approach to Topic Model with Time Factor and Clustering Methods

Many methods of discovering social networking communities or clustering of features are based on the network structure or the content network. This paper proposes a community discovery method based on topic models using a time factor and an unsupervised clustering method. Online community discovery enables organizations and businesses to thoroughly understand the trend in users’ interests in their products and services. In addition, an insight into customer experience on social networks is a tremendous competitive advantage in this era of ecommerce and Internet development. The objective of this work is to find clusters (communities) such that each cluster’s nodes contain topics and individuals having similarities in the attribute space. In terms of social media analytics, the method seeks communities whose members have similar features. The method is experimented with and evaluated using a Vietnamese corpus of comments and messages collected on social networks and ecommerce sites in various sectors from 2016 to 2019. The experimental results demonstrate the effectiveness of the proposed method over other methods.

Novel social network community discovery method combined local distance with node rank optimization function

In view of that most of the current community discovery methods in social network do not consider node self-transfer and node bias, so that it is not possible to extract the graph features effectively, which leads to the ineffective problem by the community discovery, this paper proposes a novel social network community discovery algorithm (Local Distance Laplace, LDL). First, a Laplace matrix decomposition model is constructed based on the principle of matrix decomposition. Secondly, considering the high cost of global social network information acquisition and calculation, a community discovery model based on local distance is proposed. Finally, the optimal community structure is selected by using the NRO (Node Rank Optimization) function. A comprehensive comparative analysis is made on eleven real and synthetic networks. At the same time, validation analysis is conducted on eleven different social networks (Karate, Dolphins, Lemis, Public book, Football, Celegansnertal, Email, Public blogs, Netscience, Power, Hep_th). The experimental simulation results show that: in the real network, the proposed LDL algorithm improved the overall performance by nearly 7% compared with the seven state of art optimal methods (CoVeC, EDBC, JNMF, EADP, LPANNI, LSA, SCFS). The novel algorithm is reasonable and effective and it can also be extended to multi-scale community discovery.

Modeling an web community discovery method with web page attraction

An improved Web community discovery algorithm is proposed in this paper based on the attraction between Web pages to effectively reduce the complexity of Web community discovery. The proposed algorithm treats each Web page in the Web pages collection as an individual with attraction based on the theory of universal gravitation, elaborates the discovery and evolution process of Web community from a Web page in the Web pages collection, defines the priority rules of Web community size and Web page similarity, and gives the calculation formula of the change in Web page similarity. Finally, an experimental platform is built to analyze the specific discovery process of the Web community in detail, and the changes in cumulative distribution of Web page similarity are discussed. The results show that the change in the similarity of a new page satisfies the power-law distribution, and the similarity of a new page is proportional to the size of Web community that the new page chooses to join.

Evolutionary community discovery in dynamic social networks via resistance distance

Traditional social community discovery methods concentrate mainly on static social networks, but the analysis of dynamic networks is a prerequisite for real-time and personalized social services. Through the study of community changes, the community structure in a dynamic network can be tracked over time, which helps in the mining of dynamic network information. In this paper, we propose a method of tracking dynamic community evolution that is based on resistance distance. Specifically, we model the time-varying features of dynamic networks using the convergence of a resistance-based distance. In our model, the heterogeneity of neighboring nodes can be obtained in the local topology of nodes by analyzing the resistance distance between nodes. We design a community discovery algorithm that essentially discovers community structures on dynamic networks by identifying the so-called core node. During the process of community evolution analysis, both the dynamic contribution of ordinary nodes and core nodes in each community are considered. In addition, to avoid the inclusion of spurious communities in the community structure, we define the notion of noise community and account for it in our algorithm. Experimental results show that the method proposed in this paper can yield better accuracy than other existing methods.

Overlapping Community Discovery Method Based on Two Expansions of Seeds

The real world can be characterized as a complex network sto in symmetric matrix. Community discovery (or community detection) can effectively reveal the common features of network groups. The communities are overlapping since, in fact, one thing often belongs to multiple categories. Hence, overlapping community discovery has become a new research hotspot. Since the results of the existing community discovery algorithms are not robust enough, this paper proposes an effective algorithm, named Two Expansions of Seeds (TES). TES adopts the topological feature of network nodes to find the local maximum nodes as the seeds which are based on the gravitational degree, which makes the community discovery robust. Then, the seeds are expanded by the greedy strategy based on the fitness function, and the community cleaning strategy is employed to avoid the nodes with negative fitness so as to improve the accuracy of community discovery. After that, the gravitational degree is used to expand the communities for the second time. Thus, all nodes in the network belong to at least one community. Finally, we calculate the distance between the communities and merge similar communities to obtain a less- undant community structure. Experimental results demonstrate that our algorithm outperforms other state-of-the-art algorithms.

An Online Dynamic Social Network Evolution Community Discovery Method

Online social network in the study, dynamic implicit community or group structure of discovery and detection is a very key question at the heart of evolution, it is in the medium (Mesoscopic) view to observe the online social network hidden structure characteristic, predict the evolution trend, control of the situation, found that network abnormal mass incidents, etc. It is of great significance. The purpose of this paper is to provide technical advice for community discovery research by studying online dynamic social network evolution community discovery methods. This paper analyzes the characteristics of local clustering coefficient and node similarity calculation in unsigned network, and proposes the extended local clustering coefficient as the structural attribute of the edge in unsigned network. This property can better reflect the characteristics of local network density and network structure. Combining the new edge structure measure with the label propagation algorithm with linear time complexity, a label propagation algorithm combining the extended local clustering coefficient is proposed. The results showed that the accuracy of community discovery was improved by 32.6 percent in dynamic social networks.

Multiobjective particle swarm community discovery arithmetic based on representation learning

SummaryWith the continuous increase of the network scale, the structure of the network has also become complicated. The original community discovery algorithm based on small‐scale static networks has been unable to meet our needs. In order to improve the quality of community division, community discovery algorithms based on multiple optimization functions have been proposed. These multiobjective algorithms have continued to increase in time complexity as the optimization functions increase. The time complexity of the multiobjective community discovery algorithm is reduced, and the particle swarm algorithm has higher efficiency and accuracy in solving multiobjective optimization (MOO) problems. Based on the above background, the purpose of this article is to study a multiobjective particle swarm community discovery algorithm based on representation learning. This article uses the network representation learning method for static network community discovery, and designs an improved multiobjective particle swarm‐based community discovery algorithm (MOPSO‐CD). This randomness effectively prevents the algorithm from falling into a local optimum. At the same time, combined with the MOO algorithm, all the Pareto optimal solution sets are retained to adjust the population to correct the lack of accuracy caused by the randomness of the algorithm. In addition, in order to improve the efficiency of the algorithm, this article introduces an efficient Pareto optimal solution set method. Compared with the traditional MOO strategy, the time complexity of the MOO process is O(n2) Reduced to O(nlogn). Through experimental analysis, MOPSO‐CD has higher efficiency and community discovery quality.

Finding Communities by Decomposing and Embedding Heterogeneous Information Network

Community discovery is an important task in social network analysis. However, most existing methods for community discovery rely on the topological structure alone. These methods ignore the rich information available in the content data. In order to solve this issue, in this paper, we present a community discovery method based on heterogeneous information network decomposition and embedding. Unlike traditional methods, our method takes into account topology, node content and edge content, which can supply abundant evidence for community discovery. First, an embedding-based similarity evaluation method is proposed, which decomposes the heterogeneous information network into several subnetworks, and extracts their potential deep representation to evaluate the similarities between nodes. Second, a bottom-up community discovery algorithm is proposed. Via leader nodes selection, initial community generation, and community expansion, communities can be found more efficiently. Third, some incremental maintenance strategies for the changes of networks are proposed. We conduct experimental studies based on three real-world social networks. Experiments demonstrate the effectiveness and the efficiency of our proposed method. Compared with the traditional methods, our method improves normalized mutual information (NMI) and the modularity by an average of 12% and 37% respectively.