Egocentric online social networks: Analysis of key features and prediction of tie strength in Facebook (Articolo in rivista)

Type
Label
  • Egocentric online social networks: Analysis of key features and prediction of tie strength in Facebook (Articolo in rivista) (literal)
Anno
  • 2013-01-01T00:00:00+01:00 (literal)
Alternative label
  • Valerio Arnaboldi, Andrea Guazzini, Andrea Passarella (2013)
    Egocentric online social networks: Analysis of key features and prediction of tie strength in Facebook
    in Computer communications; Elsevier, Amsterdam (Paesi Bassi)
    (literal)
Http://www.cnr.it/ontology/cnr/pubblicazioni.owl#autori
  • Valerio Arnaboldi, Andrea Guazzini, Andrea Passarella (literal)
Pagina inizio
  • 1130 (literal)
Pagina fine
  • 1144 (literal)
Http://www.cnr.it/ontology/cnr/pubblicazioni.owl#numeroVolume
  • 36 (literal)
Rivista
Http://www.cnr.it/ontology/cnr/pubblicazioni.owl#numeroFascicolo
  • 10-11 (literal)
Http://www.cnr.it/ontology/cnr/pubblicazioni.owl#affiliazioni
  • IIT-CNR (literal)
Titolo
  • Egocentric online social networks: Analysis of key features and prediction of tie strength in Facebook (literal)
Abstract
  • The widespread use of online social networks, such as Facebook and Twitter, is generating a growing amount of accessible data concerning social relationships. The aim of this work is twofold. First, we present a detailed analysis of a real Facebook data set aimed at characterising the properties of human social relationships in online environments. We find that certain properties of online social networks appear to be similar to those found \"offline\" (i.e., on human social networks maintained without the use of social networking sites). Our experimental results indicate that on Facebook there is a limited number of social relationships an individual can actively maintain and this number is close to the well-known Dunbar's number (150) found in offline social networks. Second, we also present a number of linear models that predict tie strength (the key figure to quantitatively represent the importance of social relationships) from a reduced set of observable Facebook variables. Specifically, we are able to predict with good accuracy (i.e., higher than 80%) the strength of social ties by exploiting only four variables describing different aspects of users interaction on Facebook. We find that the recency of contact between individuals - used in other studies as the unique estimator of tie strength - has the highest relevance in the prediction of tie strength. Nevertheless, using it in combination with other observable quantities, such as indices about the social similarity between people, can lead to more accurate predictions (literal)
Editore
Prodotto di
Autore CNR
Insieme di parole chiave

Incoming links:


Prodotto
Autore CNR di
Http://www.cnr.it/ontology/cnr/pubblicazioni.owl#rivistaDi
Editore di
Insieme di parole chiave di
data.CNR.it