Compactness Preserving Community Computation Via a Network Generative Process

A density based link clustering algorithm for overlapping community detection in networks

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.

One of the most significant attributes of networks is their community structure. One can define a quality criterion for a community of a network in such a way that a high-quality community refers to a group of vertices that contains more highly-connected edges between its members than between its members and the rest of the network. Various algorithms have been developed to discover communities of networks, although there is not a straightforward solution to see which algorithms are good and how they are good. However, generating synthetic graphs, which are similar to real-world networks and also have a built-in community structure (known community structure), can be used in evaluating accurately different community detection algorithms. This paper presents and programs two very important categories of synthetic networks with known community structures. Afterward, it challenges several well-known community detection algorithms to discover communities in these networks.

Community detection in complex networks via adapted Kuramoto dynamics

A cascade information diffusion based label propagation algorithm for community detection in dynamic social networks

Community detection is an effective approach to unveil relationships among individuals in online social networks. In the literature, quite a few algorithms have been proposed to conduct community detection by exploiting the topology of social networks and the attributes of social actors. In practice, community detection is usually conducted by third parties, such as advertisement companies and hospitals, with access to social networks for different purposes, which can easily lead to a privacy breach. In this paper, we investigate community detection in social networks aiming to protect the privacy of both the network topology and the users' attributes. We show that with additional prior knowledge, community detection can be performed by querying the information of only a fraction of instead of the entire population. In particular, we first propose a new scheme called differentially private community detection (DPCD). DPCD detects communities in social networks via a probabilistic generative model, which can be decomposed into subproblems solved by individual users. The private social relationships and attributes of each user are protected by objective perturbation with differential privacy guarantees. Then, we propose a parsimonious node affiliation recovery (NAR) algorithm, which is also differentially private, to unveil the community affiliation information of the whole population based on that of the limited number of queried individuals by solving a sparse optimization problem. Through both theoretical analysis and experimental validation using synthetic and real-world social networks, we demonstrate that the proposed DPCD scheme detects social communities under the modest privacy budget. In addition, we show the effectiveness of NAR to perform community detection by querying a limited number of individuals in social networks.

Community detection and tracking in social network is an important research area for many applications which are widely applied in complex systems. Recently there has been a surge of investigation in this area, fueled largely by interest in social networks, but also by interest in bibliographic citations and telecommunication records. However, due to the computational cost of the traditional algorithm for large-scale networks, most of them are not applied to industrial applications. In this paper, we present a novel framework, as CommTrend, for uncovering the community structure of networks. Our method not only has a prominent advantage over the efficiency but also can effectively reveal the underlying community structures. With respect to real-world application, our algorithm is applied to several massive datasets. Moreover, we integrate CommTrend into a bibliographical service system. By employing it to a large-scale bibliographical dataset, we demonstrate the comprehension of our work and its effectiveness for practical problems.

Community detection in social networks is a promising research area. However, network exploration with multiple types of relationships remains an underexplored yet crucial domain. To address the challenge of community detection in multiplex networks, we propose SFCSA, a smart flattening-based Cuckoo Search Algorithm, designed to address the complexities of such networks effectively. SFCSA unfolds in three integral steps. Firstly, we employ DeepWalk and node2vec algorithms to extract node embeddings, which are merged to construct the multiplex network node embeddings, minimizing information loss. Secondly, a flattened network is built by merging the network layers and adopting the Pearson correlation coefficient as a weight to capture node similarity. Thirdly, we employ a community detection approach that unveils the community structure through a cuckoo search algorithm applied to the multiplex node embeddings to extract the best set of cluster centres by maximizing the modularity function. Experimental evaluations across synthetic and real-world multiplex networks attest to the efficiency and effectiveness of our approach, outperforming state-of-the-art algorithms.

Network analysis involves using graph theory to understand networks. This knowledge is valuable across various disciplines like marketing, management, epidemiology, homeland security, and psychology. An essential task within network analysis is deciphering the structure of complex networks including technological, informational, biological, and social networks. Understanding this structure is crucial for comprehending network performance and organization, shedding light on their underlying structure and potential functions. Community structure detection aims to identify clusters of nodes with high internal link density and low external link density. While there has been extensive research on community structure detection in single‐layer networks, the development of methods for detecting community structure in multilayer networks is still in its nascent stages. In this paper, a new method, namely, IGA‐MCD, has been proposed to tackle the problem of community structure detection in multiplex networks. IGA‐MCD consists of two general phases: flattening and community structure detection. In the flattening phase, the input multiplex network is converted to a weighted monoplex network. In the community structure detection phase, the community structure of the resulting weighted monoplex network is determined using the Improved Genetic Algorithm (IGA). The main aspects that differentiate IGA from other algorithms presented in the literature are as follows: (a) instead of randomly generating the initial population, it is smartly generated using the concept of diffusion. This makes the algorithm converge faster. (b) A dedicated local search is employed at the end of each cycle of the algorithm. This causes the algorithm to come up with better new solutions around the currently found solutions. (c) In the algorithm process, chaotic numbers are used instead of random numbers. This ensures that the diversity of the population is preserved, and the algorithm does not get stuck in the local optimum. Experiments on the various benchmark networks indicate that IGA‐MCD outperforms state‐of‐the‐art algorithms.

A spreading activation-based label propagation algorithm for overlapping community detection in dynamic social networks

Robustness and Optimization of Complex Networks: Reconstructability, Algorithms and Modeling

With the rapid development of computer technology, the research on complex networks has attracted more and more attention. At present, the research directions of cloud computing, big data, internet of vehicles, and distributed systems with very high attention are all based on complex networks. Community structure detection is a very important and meaningful research hotspot in complex networks. It is a difficult task to quickly and accurately divide the community structure and run it on large-scale networks. In this paper, we put forward a new community detection approach based on internode attraction, named IACD. This algorithm starts from the perspective of the important nodes of the complex network and refers to the gravitational relationship between two objects in physics to represent the forces between nodes in the network dataset, and then perform community detection. Through experiments on a large number of real-world datasets and synthetic networks, it is shown that the IACD algorithm can quickly and accurately divide the community structure, and it is superior to some classic algorithms and recently proposed algorithms.

Detecting and analyzing community structures is a complex process in social networks. Existing methods of community detection in social networks do not consider both the contents and network structure simultaneously. Users and posts are significant subjects, as they provide wealthy information about social network characteristics, including topics that are of interest to users (topic information) and interactions between users (structure information). Based on the classical Community Overlap Propagation Algorithm (COPRA), we propose an improved Community Overlap Propagation Algorithm based on the Topics and Structure (COPRA-TS), which exploits a comprehensive similarity, and avoids randomness during the label information update process. The proposed COPRA-TS algorithm efficiently filters low-quality data, detects topics of users, computes the topic and structure similarity and utilizes an optimized label information update strategy. Experiments conducted on diverse networks demonstrate that our proposed COPRA-TS algorithm delivers high quality community detection than three notable community detection algorithms.

Community detection in complex networks has attracted a lot of attentions in recent years. Compared with the traditional single-objective community detection approaches, the multi-objective approaches based on evolutionary computation can provide a decision maker with more flexible and promising solutions. How to make effective use of the optimal solution set returned by the multi-objective community detection approaches is an important yet unsolved issue. Through leveraging an existing multi-objective community detection algorithm, this paper proposes four model selection methods to aid the decision makers to select the preferable community structures. The experiments with three synthetic and real social networks illustrate that the proposed method can discover more authentic and comprehensive community structures than traditional single-objective approaches.

In this work a brief overview of several existing approaches to find community structure and their comparison has been done. The best suited algorithm is selected on various grounds such as space and time complexities but sometimes they are not the only factors to decide whether the algorithm is perfect or not. So many factors are responsible for choosing the algorithm which are uncertain. So which algorithm is the best is still an open question? However various changes are suggested to the approaches to optimize them from time to time. Among various community structure, social network, ecological network and neural network are the most popular and currently it is being implemented on several social networking websites like twitter, LinkedIn, Face book, Amazon etc.