BDBComp
Parceria:
SBC
Diogo PachecoFernando Buarque de Lima NetoLuis MoyanoRonaldo Menezes

On the use of fully convolutional networks on evaluation of infrared breast image segmentations

In the last few years, Twitter has been used to understand several real world events such as political elections, civil conflicts, NFL games, to name a few. Motivated by these many examples we asked ourselves, what does Twitter reveal about the 2014 FIFA TM World Cup? In this paper we (i) analyze the frequency of tweets to show how support evolves during the tournament, to distinguish traditional teams from underdogs, and to show that we can identify important events during matches. Moreover, (ii) we applied clustering techniques to discover similarities on supporters behavior (tweets) that allowed us to infer countries' supporters worldwide. Lastly, (iii) for each match we created cartograms to express the volume and the level of polarization of supporters.

http://www.lbd.dcc.ufmg.br/colecoes/brasnam/2015/005.pdf

Caso o link acima esteja inválido, faça uma busca pelo texto completo na Web: Buscar na Web

Biblioteca Digital Brasileira de Computação - Contato: bdbcomp@lbd.dcc.ufmg.br
     Mantida por:
LBD