Caracterização das redes de relacionamento em sistemas de compartilhamento de fotos

• Ano de defesa: 2008
• Autor(a) principal: Daniel Coutinho de Miranda
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; UFMG
• Palavras-chave em Português: redes relacionamentos; Computação; Fotografia técnicas digitais; Processamento de imagens Técnicas digitais; Internet (Redes de computação) Aspectos sociais
• Link de acesso: http://hdl.handle.net/1843/RVMR-7L6L5F

With the increase of the popularity of the Web 2.0, several applications have encouraged more interaction between users, resulting in the formation of complex networks of relationships. The characteristics of these networks can be better understood in order to identify, for example, how users make use of the system, allowing future improvement and the proposal of new features. Example of this kind of application are photo sharing systems, which are becoming more popular due to the easiness of creating and storing photos and digital images, reinforcing the need of a more deep investigation of these systems. This dissertation presents a characterization of two types of networks of interactions between users in Flickr, a very popular Web 2.0 system to store, organize, search and share photos. The first network consists of Flickr's users interconnected by contact or friendship relations while the second network includes users interconnected by the testimonial relations. The networks were characterized in two different levels. The first level is related to the individual characteristics of the users existing in each network, aiming to identify and contrast how the system is used by the users in each network. The second level consists of the analysis of social characteristics of each network and has the goal of providing a more solid understanding of the existing interactions among the users of the networks. Our results show that users who use the testimonial feature are in general more active and more committed to using the system. Moreover, an analysis of the traffic of the photos owned by users present in both networks pointed out, among other results, that the popularity of a user photos is strongly correlated with the number of testimonials received by users and with the number of times that the user was added into contact lists. Finally, we verified that, while the contact network has strong social characteristics, the testimonial network is less social and probably reflects interactions that are based on the interest of a user by the content of others. These results were consistent across three different data sets, collected over a six month period.