Análise do impacto da topologia e das estatísticas da rede na velocidade de difusão de informações em redes sociais

• Autor(a) principal: Carvalho, André Silva de
• Data de Publicação: 2022
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações da ESPM
• Texto Completo: http://tede2.espm.br/handle/tede/691
• topic: teoria de redes; topologias de redes; estatística dos nós; contágio; difusão de informações; network theory; network topologies; node statistics; contagion; dissemination of information; CIENCIAS SOCIAIS APLICADAS::COMUNICACAO

The evolution of technologies leads to a faster diffusion of information in all contexts. Understanding the effect of information dissemination requires not only the identification of the elements that influence this process, but also the impact of different network configurations. The so-called "contagion process", in which information is transmitted by one element of a network and assimilated by another element of the same network, can undergo an acceleration due to the configuration of the space (topology) of the network or the specific characteristics of the elements that form the network, in addition to other factors. The objective of this research was to study the impact of topology and network statistics on the process of information dissemination. This was done by building computer simulations in code developed in the python language, using Microsoft's Azure cloud computing as an environment to run the simulations. The novelty of this research was to analyze the process of information diffusion considering the characteristics of theoretical networks and the real network of Facebook. Random or random networks were considered; small-world networks; free-scale networks; and topologies of real networks created from Facebook data. As a result it was observed that there is a difference between the time of contagion considering the different topologies; being the actual topology of Facebook the one with the shortest diffusion time. Still, there is no observed evidence that node statistics are sufficient, regardless of topology, to explain the time of diffusion of information in the network.