Estimation in a Binomial Stochastic Blockmodel for a Weighted Graph by a Variational Expectation Maximization Algorithm
Autor: | Abir El Haj, Zaher Khraibani, Pierre-Yves Louis, Yousri Slaoui |
---|---|
Přispěvatelé: | Laboratoire de Mathématiques et Applications (LMA-Poitiers), Université de Poitiers-Centre National de la Recherche Scientifique (CNRS), UL - Université Libanaise, Faculté des Sciences Section (1) Hadath-Beyrouth, Université Libanaise, Faculté des Sciences Section (1) Hadath-Beyrouth (UL) |
Jazyk: | angličtina |
Rok vydání: | 2020 |
Předmět: |
Statistics and Probability
050402 sociology Theoretical computer science Text mining Weighted networks 01 natural sciences Clustering 010104 statistics & probability Statistics::Machine Learning 0504 sociology [MATH.MATH-ST]Mathematics [math]/Statistics [math.ST] Statistics Expectation–maximization algorithm Poisson stochastic blockmodel 0101 mathematics Cluster analysis Mathematics Random graph [STAT.AP]Statistics [stat]/Applications [stat.AP] 05 social sciences Community structure Probabilistic logic Computer Science::Social and Information Networks Modeling and Simulation Binomial stochastic blockmodel Graph (abstract data type) Variational inference |
Zdroj: | Communications in Statistics-Simulation and Computation Communications in Statistics-Simulation and Computation, Taylor & Francis, 2020, ⟨10.1080/03610918.2020.1743858⟩ |
ISSN: | 0361-0918 1532-4141 |
DOI: | 10.1080/03610918.2020.1743858⟩ |
Popis: | International audience; Stochastic blockmodels have been widely proposed as a probabilistic random graph model for the analysis of networks data as well as for detecting community structure in these networks. In a number of real-world networks, not all ties among nodes have the same weight. Ties among networks nodes are often associated with weights that differentiate them in terms of their strength, intensity, or capacity. In this paper, we provide an inference method through a variational expectation maximization algorithm to estimate the parameters in binomial stochastic blockmodels for weighted networks. To prove the validity of the method and to highlight its main features, we set some applications of the proposed approach by using some simulated data and then some real data sets. Stochastic blockmodels belong to latent classes models. Classes defines a node's clustering. We compare the clustering found through binomial stochastic blockmodels with the ones found fitting a stochastic blockmodel with Poisson distributed edges. Inferred Poisson and binomial stochastic blockmodels mainly differs. Moreover, in our examples, the statistical error is lower for binomial stochastic blockmodels. |
Databáze: | OpenAIRE |
Externí odkaz: |