Social influence in online social network groups

Solo disponible en BuenasTareas
  • Páginas : 21 (5208 palabras )
  • Descarga(s) : 0
  • Publicado : 11 de noviembre de 2011
Leer documento completo
Vista previa del texto
Groupthink and Peer Pressure: Social Influence in Online Social Network Groups
Pan Hui, Sonja Buchegger Deutsche Telekom Laboratories, TU-Berlin Ernst-Reuter-Platz 7, D-10587 Berlin, Germany phone: +49/30/835358517, fax: +49/30/835358409 firstname.lastname@telekom.de

Abstract—In this paper, we present a horizontal view of social influence, more specifically a quantitative study of the influence ofneighbours on the probability of a particular node to join a group, on four popular Online Social Networks (OSNs), namely Orkut, YouTube, LiveJournal, and Flickr. Neighbours in OSNs have a mutually acknowledged relation, most often defined as friendship, and they are directly connected on a graph of a social network. Users in OSNs can also join groups of users. These groups represent common areasof interest. We present a simple social influence model to describe and explain the group joining process of users on OSNs. To this end, we extract the social influence from data sets of OSNs of a million sample nodes. One of our findings is that a set of neighbours in the OSN is about 100 times more powerful in influencing a user to join a group than the same number of strangers. Keywords-socialnetworks; social influence;

I. I NTRODUCTION Social influence occurs when an individual’s thoughts or actions are affected by other people. The process of interpersonal influence that affects actors’ attitudes and opinions is an important foundation of their socialization, identity, and decisions. [1]. Online social networks (OSNs), such as Facebook, Orkut, Flickr, and LiveJournal, have become verypopular and continue to grow their user base. Users on these OSNs usually have explicitly labeled friends, which we consider to be proximity relationships on the networks. Users can choose to create or join a group or community based on common interests, such as affiliations, hobbies, political stands, or religions. They can invite other users to join the same groups. It is commonly believed that thecloseproximity peers, such as friends and friends of friends, have social influence on the joining of a certain group of that node. As in any social networks, online or offline, the effect can also be the other way around, when people get to know others by virtue of their common interests and memberships in the same groups. These acquaintances then become new social relations. In OSNs, these newrelations may result in links between users and thus new neighbours in the network topology. In this paper, we analyse snapshots of network topologies. There is therefore no information on what came first, the friendship link or the group membership. At least for teenage users, it has been found that the dominant usage

pattern is to connect with friends, family, and acquaintances, therebyreinforcing the structural dimensions of peer social worlds that exist in schools. It is less common for teens to seek out new friends online [2]. We therefore think it is reasonable to assume that most users that are friends with others who joined the same group, are friends first and then join OSN groups. Group memberships have been studied in social network analysis as two-mode social networks [3],variously referred to as affiliation networks, membership networks or hypernetworks, with relations being termed affiliation relation or involvement relation. Affiliation networks exhibit a duality of social relations and affiliations. They are two-mode networks consisting of subsets of actors, and connections among members of one of the modes are based on linkages established through the second mode.More precisely, the first mode is a set of actors as usual in social networks, the second, additional mode is a set of events, which can be a wide range of specific kinds of social occasions: e.g., social clubs, boards of directors of corporations, university committees. In this paper we use the terms groups and memberships when referring to the common-interest (second mode) part of the affiliation...
tracking img