Loading...
Publication
Métricas de centralidade em redes sociais
| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 1.14 MB | Adobe PDF |
Advisor(s)
Abstract(s)
Nos modelos de redes sociais, tal como na teoria de grafos, os vértices representam os atores e as arestas ou arcos a relação entre eles. Atores influentes são aqueles que estão frequentemente envolvidos na relação com outros atores. Este envolvimento torna-os mais visíveis sendo considerados mais centrais na rede. É neste sentido que as métricas de centralidade tentam descrever as propriedades da localização de um nó fulcral numa rede. Estas medidas têm em consideração os diferentes modos de interação e comunicação de um ator com os restantes elementos, sendo mais importantes, ou centrais, aqueles que estão localizados em posições mais estratégicas na rede. Neste trabalho apresenta-se o estudo de cinco métricas de centralidade: grau, proximidade, intermediação, vetor próprio e katz. Descrevem-se os algoritmos implementados no cálculo das medidas e apresenta-se um caso de estudo. Para completar o estudo é apresentada uma análise comparativa entre os resultados obtidos no aplicativo NodeXL, e os resultados obtidos através dos algoritmos implementados.
Considering models for social networks as graphs, nodes represent the actors and the edges represent the relationship between them. Influential actors are the ones that are frequently involved on relationships between other actors. This involvement makes them more visible and considered more central on the network. In this sense centrality metrics try to describe the localization properties of an important node of the network. These measures have in consideration the different interaction and communication modes an actor has with others, being more important or central the ones that are located on more strategic locations on the network. On this work it is presented the study of five centrality measures: degree, closeness, betweenness, eigenvector and katz. It is made a description of the algorithms implemented, and it is presented a case study. To complete the study it is also made a comparative analysis between results obtained with NodeXL, and the results from the algorithms implemented.
Considering models for social networks as graphs, nodes represent the actors and the edges represent the relationship between them. Influential actors are the ones that are frequently involved on relationships between other actors. This involvement makes them more visible and considered more central on the network. In this sense centrality metrics try to describe the localization properties of an important node of the network. These measures have in consideration the different interaction and communication modes an actor has with others, being more important or central the ones that are located on more strategic locations on the network. On this work it is presented the study of five centrality measures: degree, closeness, betweenness, eigenvector and katz. It is made a description of the algorithms implemented, and it is presented a case study. To complete the study it is also made a comparative analysis between results obtained with NodeXL, and the results from the algorithms implemented.
Description
Keywords
Redes sociais Centralidade Proximidade Intermediação Vetor próprio Katz Social network Centrality Closeness Betweenness Eigenvector
Pedagogical Context
Citation
Laranjeira, Paula Alexandra; Cavique, Luís - Métricas de centralidade em redes sociais. "Revista de Ciências da Computação" [Em linha]. ISSN 1646-6330 (Print) 2182-1801 (Online). Vol. 9, nº 9 (2014), p. 1-20