Detecting professional versus personal closeness using an enterprise social network site

Abstract

In this work we analyze the behavior on a company-internal social network site to determine which interaction patterns signal closeness between colleagues. Regression analysis suggests that employee behavior on social network sites (SNSs) reveals information about both professional and personal closeness. While some factors are predictive of general closeness (e.g. content recommendations), other factors signal that employees feel personal closeness towards their colleagues, but not professional closeness (e.g. mutual profile commenting). This analysis contributes to our understanding of how SNS behavior reflects relationship multiplexity: the multiple facets of our relationships with SNS connections.

Exploring the Virtual Space of Academia

“…Sharing and obtaining information are reported as main functional drivers for joining and contributing to professional social networks [20]. This source of motivation can be enhanced by individual predispositions and organisational culture which foster a perception of knowledge as a public good, belonging not to individuals, but to the whole organization [1].…”

Enterprise social network as a social pillar in enterprise

“…Bylo zjištěno, že zpětná vazba ve formě přidaných komentářů je vysoce korelující s následnou účastí uživatele. Pracovníci IBM zkoumali detekci profesionální vs. osobní interakce v rámci podnikové sociální sítě [19]. Autoři analyzovali chování na firemní sociální síti a určili, které vzory interakce signalizují blízkost mezi kolegy.…”

Personalized Techniques for Lifestyle Change

“…Gilbert and Karaholios developed a model, which quantifies the strength of relationship between users based on 74 factors [4]. Wu et al developed another model for computing professional, personal, and overall closeness of users of an enterprise social network based on 53 factors [14]. Freyne et al developed a system for recommending social network activities based on long-and short-term models of content viewed and activities performed by users [3].…”

“…To compute the user-to-user relevance score S U (T,u x ), we adopt the model of [14] and use four categories of factors: (1) user factors (UF) -online behaviour and activity of the target user, (2) subject user factors (SUF) -online behaviour and activity of the subject user, (3) The frequency of performing actions is the main indicator of user-to-action relevance scoring. We denote by f(T,a z ) the frequency of user T performing action a z , by f(T) the average frequency of all actions performed by T, by f(a z ) the average frequency of all users performing a z , and by f() the average frequency of all actions performed by all users.…”