Social network communities’ search model
DOI:
https://doi.org/10.18372/2225-5036.26.14668Keywords:
network, security, graph, structure, communities, modeling, algorithm, verticesAbstract
Abstract. In order to ensure the smooth functioning of a social network with a large number of subscribers, it is advisable to divide it into subnets. Division into subnets will provide high-quality control of traffic and other parameters, including security parameters. The first reason for dividing the network into subnets is not to get a huge broadcast domain. The second important reason for dividing the network into subnets is to provide a certain level of security. The third reason is identifying online communities. The necessity of creating a model is determined. In this model graph is randomly generated with specified parameters for internal and external relationships between vertices, and communities are considered extraordinary. A method for isolating the structure of communities based on the maximum likelihood method is proposed, and a numerical random search algorithm is described on its basis. Graphs which representing real social and communication networks are changed rapidly. Moreover, random graphs are an effective tool for studying these networks. An important task is to identify the structure of communities in networks. In conditions of large dimensionality of networks, approximate methods are especially relevant. That allow finding a solution close to the optimal for a limited time. To solve this problem, it is proposed to create a model for distinguishing the structure of communities based on the maximum likelihood method. And a description of a numerical random search algorithm based on created model. This paper describes a mathematical model in which a graph is randomly generated with specified parameters for internal and external relationships between vertices, and communities are considered extraordinary. A method for isolating the structure of communities based on the maximum likelihood method is proposed, and a numerical random search algorithm using the Boltzmann-Gibbs distribution is described on its basis. The behavior of the objective function is investigated.
References
В. Ахрамович, "Ідентифікація й аутенти-фікація, керування доступом", Сучасний захист інфо-рмації, К.:ДУТ, №4, С. 47-51, 2016.
В. Ахрамович., В. Чегренець, "Уразливості та способи захисту бездротових мереж. Тенденції розвитку конвергентних мереж: рішення пост - NGN, 4G и 5G", ІI Міжнародна науково-практична конферен-ція, 17-18 листопада 2016 року: тези доп., К., С. 163-166, 2016.
В. Ахрамович., В. Чегренець, Інформаційна безпека [практикум], К.: ДУТ, 2017, 396 с.
В. Чегренец, В. Ахрамович, Н. Руденко, "Хмарні технології та можливості використання в комп’ютерних телекомунікаційних системах", Наукові записки Українського науково-дослідного інституту зв’ язку, №2 (42), С. 121-129, 2016.
K. Avrachenkov, A. Kondratev, V. Mazalov, "Cooperative Game Theory approaches for network partitioning", Computing and Combinatorics. Eds. by Y. Cao, J. Chen. COCOON, 2017, LNCS, 2017, vol. 10392, pp. 591–603.
J. Copic, M. Jackson, A. Kirman, "Identifying community structures from network data via maximum likelihood methods", The B. E. Journal of Theoretical Economics, vol. 9, iss. 1, pp. 1935–1704, 2009.
M. Jackson, Social and economic networks. Princeton, Princeton University Press, 2010, 520 p.
R. Kaur, S. Singh, "A survey of data mining and social network analysis based anomaly detection techniques", Egypt Inf. Journal, vol. 17(2), pp. 199–216, 2016.
M. Newman, M. Girvan, "Finding and evaluating community structure in networks", Physical Review E, vol. 69(2), pp. 26-53, 2004.
M. Meila, J. Shi, "A Random Walks View of spectral segmentation", Proceedings of AISTATS, pp. 1–6, 2001.
R. Myerson, "Graphs and cooperation in games", Math. Oper. Res., vol. 2, pp. 225–229, 1977.
