Download or read book Community Detection in Networks with Node Covariates written by Zahra Razaee and published by . This book was released on 2017 with total page 93 pages. Available in PDF, EPUB and Kindle. Book excerpt: Community detection or clustering is a fundamental task in the analysis of network data. Most networks come with annotations which can be in form of node covariates such as a person's age, gender and location and/or edge covariates such as time stamps and ratings. However, most of the existing community detection approaches infer the community memberships merely based on the network structure. Moreover, many real networks have a bipartite structure which makes community detection challenging. In this dissertation, we first propose a model-based approach which allows for matched communities in the bipartite setting, in addition to node covariates with information about the matching. We derive a simple fast algorithm for fitting the model, based on variational inference ideas. A variation of the model to allow for degree-correction is also considered, in addition to a novel approach to fitting such degree-corrected models. We also propose a unified affinity matrix (USim) to leverage the node covariates information that can be used in unipartite networks (directed and undirected) as well as the bipartite networks that combines the information from the network with that from the node covariates into a single similarity matrix, which can then be input to a spectral clustering algorithm. We show the effectiveness of both approaches on simulated and real data, namely, page-user networks collected from Wikipedia.
Download or read book Community Detection in Social Networks written by Sihan Huang and published by . This book was released on 2020 with total page pages. Available in PDF, EPUB and Kindle. Book excerpt:
Download or read book Machine Learning Methods for Community Detection in Networks Using Known Community Information written by Meghana Venkata Palukuri and published by . This book was released on 2022 with total page 0 pages. Available in PDF, EPUB and Kindle. Book excerpt: In a network, the problem of community detection refers to finding groups of nodes and edges that form ‘communities’ relevant to the field, such as groups of people with common interests in social networks and fraudulent websites linked to each other on the web. Community detection also yields downstream use-cases such as the summarization of massive networks into smaller networks of communities. We are most interested in mining protein complexes, i.e., communities of interacting proteins, accelerating biological experiments by providing candidates for previously unknown protein complexes. Characterization of protein complexes is important, as they play essential roles in cellular functions and their disruption often leads to disease. Previous methods in community detection comprise a majority of unsupervised graph clustering strategies, which work on the assumption that communities are dense subgraphs in a network - which is not always true. Also, many community detection algorithms are in-memory and serial and do not scale to large networks. In this dissertation, we use knowledge from communities, including rich features from graph nodes, with supervised and reinforcement learning, improving on accuracies, with parallel algorithms ensuring high performance and scalability. Specifically, we work on (1) learning a community fitness function using supervised machine learning methods with AutoML; (2) a distributed algorithm for finding candidate communities using multiple heuristics; (3) learning to walk trajectories on a network leading to communities with reinforcement learning and (4) feature augmentation with graph node information, such as images and additional graph node embeddings. While we optimize our algorithms on protein complexes that have characteristics such as being overlapping in nature with different topologies, our methods are generalizable to other domains since they learn and use characteristics of communities to predict new communities. Further, in domains with limited known information, the algorithms we develop can be applied by transferring learned knowledge such as dense community fitness functions from other domains. In conclusion, we build Super.Complex, RL complex detection, and DeepSLICEM - three accurate, efficient, scalable, and generalizable community detection algorithms, that effectively utilize known community information with different machine learning methods and also present 3 evaluation measures for accurate community evaluation
Download or read book Bipartite Network Community Detection Development and Survey of Algorithmic and Stochastic Block Model Based Methods written by Yidan Sun and published by . This book was released on 2021 with total page 144 pages. Available in PDF, EPUB and Kindle. Book excerpt: In a bipartite network, nodes are divided into two types, and edges are only allowed to connect nodes of different types. Bipartite network clustering problems aim to identify node groups with more edges between themselves and fewer edges to the rest of the network. The approaches for community detection in the bipartite network can roughly be classified into algorithmic and model-based methods. The algorithmic methods solve the problem either by greedy searches in a heuristic way or optimizing based on some criteria over all possible partitions. The model-based methods fit a generative model to the observed data and study the model in a statistically principled way. In this dissertation, we mainly focus on bipartite clustering under two scenarios: incorporation of node covariates and detection of mixed membership communities. In Chapter 2, we study an algorithmic bipartite clustering method in the context of gene co-clustering. Gene co-clustering is a widely-used technique that has enabled computational prediction of unknown gene functions within a species. However, it remains a challenge to refine gene function prediction by leveraging evolutionarily conserved genes in another species. This challenge calls for a new computational algorithm to identify gene co-clusters in two species, so that genes in each co-cluster exhibit similar expression levels in each species and strong conservation between the species. Here we develop the bipartite tight spectral clustering (BiTSC) algorithm, which identifies gene co-clusters in two species based on gene orthology information and gene expression data. BiTSC novelly implements a formulation that encodes gene orthology as a bipartite network and gene expression data as node covariates. This formulation allows BiTSC to adopt and combine the advantages of multiple unsupervised learning techniques: kernel enhancement, bipartite spectral clustering, consensus clustering, tight clustering, and hierarchical clustering. As a result, BiTSC is a flexible and robust algorithm capable of identifying informative gene co-clusters without forcing all genes into co-clusters. Another advantage of BiTSC is that it does not rely on any distributional assumptions. Beyond cross-species gene co-clustering, BiTSC also has wide applications as a general algorithm for identifying tight node co-clusters in any bipartite network with node covariates. We demonstrate the accuracy and robustness of BiTSC through comprehensive simulation studies. In a real data example, we use BiTSC to identify conserved gene co-clusters of D. melanogaster and C. elegans, and we perform a series of downstream analyses to both validate BiTSC and verify the biological significance of the identified co-clusters. The stochastic block model (SBM) serves as a fundamental and classic tool to study model-based community detection in a statistically principled way. It is a simple generative model but too restrictive to include empirical network characteristics. With the flexibility of the SBM, many SBM variants have been proposed to accommodate the real-world scenarios, such as node degree heterogeneity, mixed membership communities, and bipartite structure. In Chapter 3, we review a few extensions of the SBM, including degree-correction SBM, mixed membership SBM, and bipartite SBM. We conduct a survey about community detection methods based on the mixed membership SBM and bipartite SBM. We discuss these methods in detail and summarize the challenges in introducing degree-correction to the mixed membership SBM; the challenges in extending the degree-corrected mixed membership SBM to the bipartite network. To address some of the research gaps discussed in Chapter 3, we construct a generative bipartite degree-corrected mixed membership SBM in Chapter 4. To fit the model, we propose a variational expectation-maximization (EM) algorithm. We perform a preliminary simulation study in which the result suggests that the algorithm is sensitive to the initial values. We discuss the possible improvements for future directions.
Download or read book Network Inference with Statistical Guarantees written by Xueyu Mao and published by . This book was released on 2020 with total page 504 pages. Available in PDF, EPUB and Kindle. Book excerpt: Networks arise in a huge variety of real data scenarios: starting from social networks like Facebook or user product networks in recommendation systems to protein-protein interaction networks in biological systems, etc. In this thesis, we focus on developing fast and provable algorithms for some network inference problems. In the first part, we are interested in overlapping community detection problem under the popular Mixed Membership Stochastic Blockmodel (MMSB). We firstly establish sufficient conditions for the symmetric non-negative matrix factorization optimization to have a unique solution under MMSB, and propose a computationally efficient algorithm called GeoNMF that is provably optimal and hence consistent for a broad parameter regime. Then using the inherent geometry of MMSB, we link the inference of overlapping communities to the problem of finding corners in a noisy rotated and scaled simplex, for which consistent algorithms exist. We use this as a building block for our algorithm to infer the community memberships of each node, and provide uniform rates of convergence for the inferred community membership vector of each node in the network. As a byproduct of our analysis, we derive sharp row-wise eigenvector deviation bounds, and provide a cleaning step that improves the performance drastically for sparse networks. Our results hold over a broad parameter regime where the average degree only grows poly-logarithmically with the number of nodes. Using experiments with simulated and real datasets, we show that our method achieves better error with lower variability over competing methods, and processes real world networks of up to 100,000 nodes within tens of seconds. For the second part, we go beyond MMSB for overlapping community detection. Notice that many existing overlapping clustering methods model each person (or word, or book) as a non-negative weighted combination of "exemplars" who belong solely to one community, with some small noise. Geometrically, each person is a point on a cone whose corners are these exemplars. This basic form encompasses the widely used MMSB of networks and its degree corrected variants, as well as topic models such as LDA. We show that a simple one-class SVM yields provably consistent parameter inference for all such models, and scales to large datasets. Experimental results on several simulated and real datasets show our algorithm (called SVM-cone) is both accurate and scalable. The final contribution of this thesis is novel nonparametric methods for network covariate estimation. Networks with node covariates are commonplace: for example, people in a social network have interests, or product preferences, etc. If we know the covariates for some nodes, can we infer them for the remaining nodes? We provide two provably consistent methods to solve this problem. For "low-rank" latent variable models, we develop SVD-RBF, which uses the top principal components of the network in a non-parametric regression. For general models, we present CN-VEC, which constructs a similarity measure between two nodes, based on the patterns of their 2-hop neighborhoods. CN-VEC then predicts node covariates by averaging the covariates of the top-k most similar nodes using this measure. SVD-RBF is consistent for low-rank models when the average degree grows with Õ(log n), while CN-VEC is consistent for a wide range of models when the degree grows with Õ(n1 [superscript /] 3). To our knowledge, CN-VEC is the first provably consistent method for this problem under general models. Both methods are fast, and CN-VEC is also parameter-free. Experiments on 4 simulated network models and 3 real-world datasets show the effectiveness of our algorithms compared to the state of the art
Download or read book From Security to Community Detection in Social Networking Platforms written by Panagiotis Karampelas and published by Springer. This book was released on 2019-04-09 with total page 242 pages. Available in PDF, EPUB and Kindle. Book excerpt: This book focuses on novel and state-of-the-art scientific work in the area of detection and prediction techniques using information found generally in graphs and particularly in social networks. Community detection techniques are presented in diverse contexts and for different applications while prediction methods for structured and unstructured data are applied to a variety of fields such as financial systems, security forums, and social networks. The rest of the book focuses on graph-based techniques for data analysis such as graph clustering and edge sampling. The research presented in this volume was selected based on solid reviews from the IEEE/ACM International Conference on Advances in Social Networks, Analysis, and Mining (ASONAM '17). Chapters were then improved and extended substantially, and the final versions were rigorously reviewed and revised to meet the series standards. This book will appeal to practitioners, researchers and students in the field.
Download or read book Descriptive vs Inferential Community Detection in Networks written by Tiago P. Peixoto and published by Cambridge University Press. This book was released on 2023-08-31 with total page 146 pages. Available in PDF, EPUB and Kindle. Book excerpt: Community detection is one of the most important methodological fields of network science, and one which has attracted a significant amount of attention over the past decades. This Element closes the gap between the state-of-the-art in community detection on networks and the methods actually used in practice.
Download or read book A Survey of Statistical Network Models written by Anna Goldenberg and published by Now Publishers Inc. This book was released on 2010 with total page 118 pages. Available in PDF, EPUB and Kindle. Book excerpt: Networks are ubiquitous in science and have become a focal point for discussion in everyday life. Formal statistical models for the analysis of network data have emerged as a major topic of interest in diverse areas of study, and most of these involve a form of graphical representation. Probability models on graphs date back to 1959. Along with empirical studies in social psychology and sociology from the 1960s, these early works generated an active network community and a substantial literature in the 1970s. This effort moved into the statistical literature in the late 1970s and 1980s, and the past decade has seen a burgeoning network literature in statistical physics and computer science. The growth of the World Wide Web and the emergence of online networking communities such as Facebook, MySpace, and LinkedIn, and a host of more specialized professional network communities has intensified interest in the study of networks and network data. Our goal in this review is to provide the reader with an entry point to this burgeoning literature. We begin with an overview of the historical development of statistical network modeling and then we introduce a number of examples that have been studied in the network literature. Our subsequent discussion focuses on a number of prominent static and dynamic network models and their interconnections. We emphasize formal model descriptions, and pay special attention to the interpretation of parameters and their estimation. We end with a description of some open problems and challenges for machine learning and statistics.
Download or read book Fuzzy Systems and Data Mining IX written by A.J. Tallón-Ballesteros and published by IOS Press. This book was released on 2023-12-19 with total page 980 pages. Available in PDF, EPUB and Kindle. Book excerpt: Fuzzy systems and data mining are indispensible aspects of the digital technology on which we now all depend. Fuzzy logic is intrinsic to applications in the electrical, chemical and engineering industries, and also in the fields of management and environmental issues. Data mining is indispensible in dealing with big data, massive data, and scalable, parallel and distributed algorithms. This book presents the proceedings of FSDM 2023, the 9th International Conference on Fuzzy Systems and Data Mining, held from 10-13 November 2023 as a hybrid event, with some participants attending in Chongqing, China, and others online. The conference focuses on four main areas: fuzzy theory, algorithms and systems; fuzzy application; data mining; and the interdisciplinary field of fuzzy logic and data mining, and provides a forum for experts, researchers, academics and representatives from industry to share the latest advances in the field of fuzzy sets and data mining. This year, topics from two special sessions on granular-ball computing and the application of generative AI, as well as machine learning and neural networks, were also covered. A total of 363 submissions were received, and after careful review by the members of the international program committee, 110 papers were accepted for presentation at the conference and publication here, representing an acceptance rate of just over 30%. Covering a comprehensive range of current research and developments in fuzzy logic and data mining, the book will be of interest to all those working in the field of data science.
Download or read book Scalable Community Detection for Social Networks written by Arnau Prat Pérez and published by . This book was released on 2016 with total page 137 pages. Available in PDF, EPUB and Kindle. Book excerpt: Many applications can be modeled intuitively as graphs, where nodes represent the entities and the edges the relationships between them. This way, we are able to better understand them and how they interact. One particularity of these graphs is that their entities are organized in modules called communities. A community is informally defined as a set of nodes more densely connected internally than externally. For instance, in the case of a social network, persons with similar characteristics are grouped forming communities. Community detection has become a hot topic in the research community during the last years, due to its amount of applications. For instance, in social networks, communities give information about the persons forming them, by just looking at the relationships linking them. This is used in directing marketing campaigns, recomendation systems or in link prediction. Because of the relevance of the problem, many community detection algorithms exist, which follow different strategies. Most of them are based on the well known modularity metric, though other techniques based on random walks and epidemics spreading also exist. The problem of existing algorithms is that they have been designed to be generic, completely ignoring the particularities of the graphs belonging to different domains. As a result and under certain circumstances, these algorithms tend to find groups of nodes with a lack of a community structure. This thesis, overcomes this issues by proposing a novel community detection algorithm design methodology, called Domain Specific Community detection. This methodology is based on defining a set of structural properties communities of a given domain should fulfill, as well a set of behavioral properties to be fulfilled by a community detection algorithm or metric. Based on this methodology, we propose a set of properties for the specific domain of social networks, consisting of three structural properties (Internal structure sensitive, Bridges resistant and Cut-Vertex resistant) and three behavioral properties (Scale independent, Adaptive and Lineal community cohesion). Based on the aforementioned properties, we design a novel community detection metric, called the Weighted Community Clustering (WCC), which takes the presence of a triangle as an indicator of a strong relation between two persons in a social network. We formally prove that WCC fulfills the proposed properties, thus guaranteeting that communities resulting from maximizing WCC have a minimum degree of quality. Moreover, we prove this last statement by performing an empirical analysis on communities from real graphs, showing that WCC is able to correclty rank these well. In this thesis we also propose an algorithm called Scalable Community Detection (SCD), based on the maximization of WCC. SCD is also designed with parallelism in mind, in order to take advantage of current many-core architectures. We show that SCD is to detect communities with an unprecedented quality, being its execution time faster than most of existing proposals, being able to process billion edge graphs in a few hours This thesis also includes a statistical study about the structural characteristics of the meta-groups found in several real graphs, comparing these to graph from two different synthetic graph generators. We show that communities produced by a synthetic graph generator commonly used in community detection research are very dissimilar to those found in real graphs. Finally, this thesis includes a study on how to implement a triangle counting algorithm on a modern many core architecture, more concretely the Intel Single Chip Cloud Computer (Intel SCC).
Download or read book Local Community Detection on Complex Networks written by Yuchen Bian and published by . This book was released on 2019 with total page pages. Available in PDF, EPUB and Kindle. Book excerpt: Community detection or clustering plays a vital role in data mining and machine learning. Recently local community detection attracts more attention due to its online manner. Given a community-structured graph and a set of query nodes (one query node or multiple query nodes), local community detection aims to detect local communities (one community or multiple communities) containing the queries. Two main assumptions exist in the literature. One is the query-bias assumption which assumes that nodes in the target communities locate near the query nodes. The other is the homology assumption which assumes that the query nodes come from the same community. However, these two assumptions prevent local community detection methods to spot target communities accurately. In this dissertation, three models are proposed to address issues resulting from the two assumptions.Suppose only one query node is given and it does not come from communities overlapping region (i.e., there is only one target community). The query-bias assumes that nodes near the query node have high chances to be detected in the target community. Usually, many existing methods send one random walker from the query to explore the graph. The probability of the walker visiting one node is the chance of that node being detected in the target community. Nevertheless, when the query is on the community boundary region, the random walker has a high probability to travel to nearby communities but less probability to visit nodes in the target community but disconnect the query. The query-bias assumption results in false positives (nodes near the query node but not in the target community will be detected) and false negatives (nodes in the target community but not near the query are missed). To avoid the query-bias, in this dissertation, a multiple walkers chain (MWC) model is proposed where a group of mutually affected random walkers is sent from the query node. Due to the interaction between walkers, the walkers group in MWC is very hard to escape from the target community. The results on synthetic networks and real networks show that MWC outperforms other existing works.Suppose multiple query nodes are given. The homology assumption considers the query nodes from the same target community. However, the homology assumption does not hold in many cases especially when different domains are taken into account. Even when the homology assumption is neglected, if there are too many query nodes, detecting the target community for each query at a time will not only suffer from the query-bias but also cost too much. If several query nodes are known in advance that they are homologous, they should be taken care of together to detect their target community. In this dissertation, a memory-based random walker (MRW) model and an interactive framework are proposed to detect multiple communities for a set of query nodes without holding the homology assumption. One random walker is sent from each query node and each walker memorizes its visiting history. During the exploring process, the interaction between walkers helps to divide the multiple query nodes into groups in the meantime of detecting. In this way, the algorithm can execute just once to detect all target communities for multiple queries. Experimental results on real-world networks show that the proposed MRW performs better and faster than the alternatives.Local community detection for query nodes from communities overlapping region becomes extremely hard for exiting methods due to the two assumptions. Most works cannot detect multiple communities for one overlapping query node. In this dissertation, a general replacement mechanism is proposed to deal with all the aforementioned three intractable query cases: boundary query, multiple queries, and overlapping query. The main idea is to replace intractable queries with detection-friendly new queries (nodes in the community core region). First, brought from the field theory in physics, a query-oriented topology potential (TP) filed is built around one query node. Then TP values of core nodes are emphasized with a designed amplifier. Finally, the query-oriented core nodes are distinguished and easy to detect. The detected core nodes are used as new queries to find the target local communities. More importantly, the core nodes detection strategy can be applied in other graph mining applications. Experimental results verify the advantages of the replacement strategy.
Download or read book Data Analysis For Network Cyber security written by Niall M Adams and published by World Scientific. This book was released on 2014-04-04 with total page 200 pages. Available in PDF, EPUB and Kindle. Book excerpt: There is increasing pressure to protect computer networks against unauthorized intrusion, and some work in this area is concerned with engineering systems that are robust to attack. However, no system can be made invulnerable. Data Analysis for Network Cyber-Security focuses on monitoring and analyzing network traffic data, with the intention of preventing, or quickly identifying, malicious activity.Such work involves the intersection of statistics, data mining and computer science. Fundamentally, network traffic is relational, embodying a link between devices. As such, graph analysis approaches are a natural candidate. However, such methods do not scale well to the demands of real problems, and the critical aspect of the timing of communications events is not accounted for in these approaches.This book gathers papers from leading researchers to provide both background to the problems and a description of cutting-edge methodology. The contributors are from diverse institutions and areas of expertise and were brought together at a workshop held at the University of Bristol in March 2013 to address the issues of network cyber security. The workshop was supported by the Heilbronn Institute for Mathematical Research.
Download or read book Community Detection in Network Analysis written by Lingjia Zhang and published by . This book was released on 2016 with total page 70 pages. Available in PDF, EPUB and Kindle. Book excerpt: The existence of community structures in networks is not unusual, including in the domains of sociology, biology, and business, etc. The characteristic of the community structure is that nodes of the same community are highly similar while on the contrary, nodes across communities present low similarity. In academia, there is a surge in research efforts on community detection in network analysis, especially in developing statistically sound methodologies for exploring, modeling, and interpreting these kind of structures and relationships. This survey paper aims to provide a brief review of current applicable statistical methodologies and approaches in a comparative manner along with metrics for evaluating graph clustering results and application using R. At the end, we provide promising future research directions.
Download or read book Detecting Shared Communities in Social Networks Algorithmically written by Himansu Sekhar Pattanayak and published by . This book was released on 2023-04-10 with total page 0 pages. Available in PDF, EPUB and Kindle. Book excerpt: Detecting shared communities in social networks algorithmically involves the use of graph theory and clustering algorithms to identify groups of individuals in a social network that are closely connected and share common interests or characteristics. The algorithmic approach to community detection can be broken down into the following steps: Graph construction: A social network can be represented as a graph where nodes represent individuals and edges represent connections between them. A graph can be constructed based on the relationships between individuals in the network.Node degree calculation: The degree of a node is the number of connections it has in the network. This information can be used to identify individuals who are highly connected and potentially part of a community.Community detection: Clustering algorithms can be applied to the graph to group nodes that are highly connected and have similar characteristics or interests. There are many algorithms that can be used for community detection, such as the Girvan-Newman algorithm, the Louvain algorithm, and the Info map algorithm.Community analysis: Once communities have been detected, various measures can be used to analyze them. For example, the modularity of a community can be calculated to measure how well the group is connected internally and how different it is from the rest of the network. Other measures, such as centrality and density, can also be used to characterize communities.Overall, the algorithmic approach to community detection in social networks can help identify shared interests, behaviors, and characteristics of individuals in a network, which can be useful for targeted marketing, social science research, and other applications.
Download or read book Theoretical Analysis for Convex and Non convex Clustering Algorithms written by Bowei Yan and published by . This book was released on 2018 with total page 382 pages. Available in PDF, EPUB and Kindle. Book excerpt: Clustering is one of the most important unsupervised learning problem in the machine learning and statistics community. Given a set of observations, the goal is to find the latent cluster assignment of the data points. The observations can be either some covariates corresponding to each data point, or the relational networks representing the affinity between pair of nodes. We study the problem of community detection in stochastic block models and clustering mixture models. The two kinds of problems bear a lot of resemblance, and similar techniques can be applied to solve them. It is common practice to assume some underlying model for the data generating process in order to analyze it properly. With some pre-defined partitions of all data points, generative models can be defined to represent those two types of data observations. For the covariates, the mixture model is one of the most flexible and widely-used models, where each cluster i comes from some distribution D [subscript i], and the entire distribution is a convex sum over all distributions [mathematical equation]. We assume that the data is Gaussian or sub-gaussian, and analyze two algorithms: 1) Expectation-Maximization algorithm, which is notoriously non-convex and sensitive to local optima, and 2) Convex relaxation of the k-means algorithm. We show both methods are consistent under certain conditions when the signal to noise ratio is relatively high. And we obtain the upper bounds for error rate if the signal to noise ration is low. When there are outliers in the data set, we show that the semi-definite relaxation exhibits more robust result compared to spectral methods. For the networks, we consider the Stochastic Block Model (SBM), in which the probability of edge presence is fully determined by the cluster assignments of the pair of nodes. We use a semi-definite programming (SDP) relaxation to learn the clustering matrix, and discuss the role of model parameters. In most SDP relaxations of SBM, the number of communities is required for the algorithm, which is a strong requirement for many real-world applications. In this thesis, we propose to introduce a regularization to the nuclear norm, which is shown to be able to exactly recover both the number of communities and cluster memberships even when the number of communities is unknown. In many real-world networks, it is more common to see both network structure and node covariates simultaneously. In this case, we present a regularization based method to effectively combine the two sources of information. The proposed method works especially well when the covariates and network contain complementary information.
Download or read book Community Detection in Large Networks written by June Andrews and published by . This book was released on 2012 with total page 81 pages. Available in PDF, EPUB and Kindle. Book excerpt: Graphs are used to represent various large and complex networks in scientific applications. In order to understand the structure of these graphs, it is useful to treat a set of nodes with similar characteristics as one community and analyze the community's behavior as a whole. Finding all such communities within the graph is the object of community detection. In our research, we compare dozens of existing community detection methods and develop a new class of algorithms for finding communities.
Download or read book Community Detection In Evolving Networks written by Tejas Puranik and published by . This book was released on 2017 with total page 117 pages. Available in PDF, EPUB and Kindle. Book excerpt: Most social networks are characterized by the presence of community structure, viz. the existence of clusters of nodes with a much higher proportion of links within the clusters than between the clusters. Community detection has many applications in many kinds of networks, including social networks and biological networks. Many different approaches have been proposed to solve the problem. An approach that has been shown to scale well to large networks is the Louvain method, based on maximizing modularity, which is a quality function of a partition of the nodes.In this thesis, we address the problem of community detection in evolving social networks. As social networks evolve, the community structure of the network can change. How can the community structure be updated in an efficient way? How often should community structure be updated? In this thesis, we give two methods based on the Louvain algorithm, to determine when to update the community structure. The first method, called the Edge-Distribution-Analysis algorithm, analyzes the newly added edges in order to make this decision. The second method, called the Modularity-Change-Rate algorithm, finds the rate of modularity change in a given network, and uses it to predict whether an update is required.Due to the sparsity of real datasets of evolving networks, we propose three models to generate evolving networks: a Random model, a model based on the well-known phenomenon of homophily in social networks, and another based on the phenomenon of triadic and cyclic closure. Starting with real-world data sets, we used these models to generate evolving networks. We evaluated the Edge-Distribution-Analysis algorithm and Modularity-Change-Rate algorithm on these data sets. Our results show that both our methods predict quite well when the community structure should be updated. They result in significant computational savings compared to approaches that would update the community structure after a fixed number of edge additions, while ensuring that the quality of the community structure is comparable.