We develop the relational topic model (RTM), a model of documents and the links between them. For each pair of documents, the RTM models their link as a binary random variable that is conditioned on their contents. The model can be used to summarize a network of documents, predict links between them, and predict words within them. We derive efficient inference and learning algorithms based on variational methods and evaluate the predictive performance of the RTM for large networks of scientific abstracts and web documents. 1

Economic Forum within the framework of the

Although a large body of work are devoted to finding communities in static social networks, only a few studies examined the dynamics of communities in evolving social networks. In this paper, we propose a dynamic stochastic block model for finding communities and their evolutions in a dynamic social network. The proposed model captures the evolution of communities by explicitly modeling the transition of community memberships for individual nodes in the network. Unlike many existing approaches for modeling social networks that estimate parameters by their most likely values (i.e., point estimation), in this study, we employ a Bayesian treatment for parameter estimation that computes the posterior distributions for all the unknown parameters. This Bayesian treatment allows us to capture the uncertainty in parameter values and therefore is more robust to data noise than point estimation. In addition, an efficient algorithm is developed for Bayesian inference to handle large sparse social networks. Extensive experimental studies based on both synthetic data and real-life data demonstrate that our model achieves higher accuracy and reveals more insights in the data than several state-of-theart algorithms.

Abstract not found

Latent structure in complex networks, e.g., in the form of community structure, can help understand network dynamics, identify heterogeneities in network properties, and predict ‘missing ’ links. While most community detection algorithms are based on optimizing heuristic clustering objectives such as the Modularity, it has recently been shown that latent structure in complex networks is learnable by Bayesian generative link distribution models (Airoldi et al., 2008, Hofman and Wiggins, 2008). In this paper we propose a new generative model that allows representation of latent community structure as in the previous Bayesian approaches and in addition allows learning of node specific link properties similar to that in the modularity objective. We employ a new relaxation method for efficient inference in these generative models that allows us to learn the behavior of very large networks. We compare the link prediction performance of the learning based approaches and other widely used link prediction approaches in 14 networks ranging from medium size to large networks with more than a million nodes. While link prediction is typically well above chance for all networks, we find that the learning based mixed membership stochastic block model of Airoldi et al., performs well and often best in our experiments. The added complexity of the LD model improves link predictions for four of the 14 networks. 1

Abstract Many algorithms have been designed to discover community structure in networks. Most of these detect disjoint communities, while a few can find communities that overlap. We propose a new, two-phase, method of detecting overlapping communities. In the first phase, a network is transformed to a new one by splitting vertices, using the idea of split betweenness; in the second phase, the transformed network is processed by a disjoint community detection algorithm. This approach has the potential to convert any disjoint community detection algorithm into an overlapping community detection algorithm. Our experiments, using several “disjoint ” algorithms, demonstrate that the method works, producing solutions, and execution times, that are often better than those produced by specialized “overlapping ” algorithms. 1

Abstract not found

In this paper, we consider the problem of community detection in directed networks by using probabilistic models. Most existing probabilistic models for community detection are either symmetric in which incoming links and outgoing links are treated equally or conditional in which only one type (i.e., either incoming or outgoing) of links is modeled. We present a probabilistic model for directed network community detection that aims to model both incoming links and outgoing links simultaneously and differentially. In particular, we introduce latent variables node productivity and node popularity to explicitly capture outgoing links and incoming links, respectively. We demonstrate the generality of the proposed framework by showing that both symmetric models and conditional models for community detection can be derived from the proposed framework as special cases, leading to better understanding of the existing models. We derive efficient EM algorithms for computing the maximum likelihood solutions to the proposed models. Extensive empirical studies verify the effectiveness of the new models as well as the insights obtained from the unified framework.

Abstract: The study of community structures is a hot spot for many inhomogeneous networks. Modularity plays an important role in this area, because it is a criterion for community detection, and a basis for community detection algorithms. Although commonly used in papers concerning community structures, modularity is seldom fully studied. In this paper, we investigate problems with the properties of modularity as defined by Newman and we propose a modularity normalized for number of groups as well as a hybrid modularity that improves on properties that reflect the interactions among communities. We also illustrate the basic flowchart of a “bottom-up merging” community detection strategy based on the properties of modularity, and explore a detection algorithm inspired by hybrid modularity. The central idea of “Community Structure ” is widely used in the study of social, biological and technical networks, among others. It represents an important future direction of complex networks research. [1-8] In order to evaluate community structure, several measurements based on network density have been developed. [9] Modularity is

This is an Open Access article distributed under the terms of the Creative Commons Attribution License