Feasibility of structural network clustering for group-based privacy control in social networks

Abstract

Users of social networking sites often want to manage the sharing of information and content with different groups of people based on their differing relationships. However, grouping contacts places a significant configuration burden on the user. Automated approaches to grouping may have the potential to reduce this burden, however, their use remains largely untested.We investigate people's rationales when grouping their contacts for the purpose of controlling their privacy, finding six criteria that they commonly considered. We assess an automated approach to grouping, based on a network clustering algorithm, whose performance may be analogous to the human's use of some of these criteria. We find that the similarity between the groups created by people and those created by the algorithm is correlated with the modularity of their network. We also demonstrate that the particular clustering algorithm, SCAN, which detects hubs and outliers within a network can be beneficial for identifying contacts who are hard to group or for whom privacy preferences are inconsistent with the rest of their group.

Friend Grouping Algorithms for Online Social Networks: Preference, Bias, and Implications

“…Early work exploring group creation focused primarily on privacy control interfaces [14,16]. In this domain, manual creation and annotation of groups was costly in terms of time and frustration due to unintuitive interfaces.…”

“…Studies on group creation demonstrate that people are not willing to use current grouping techniques in OSNs as they were intended. For example, a study on Facebook lists at 2010 showed that only 20% of participants' friends were included in Facebook lists and none of the participants used these lists for controlling privacy [14]. In a related study, Kelly et al asserted that participants using Facebook lists to create groups included few friends [16].…”

“…In 2011, Google+ introduced circles that enable selective sharing and filtering of posts on the site. Recent studies have emphasized the desire and feasibility of grouping for privacy control, sharing, and filtering [14,16,15,27]. These studies found that people desired groupings or clusters of members in their community.…”

“…Given the significant burden of manual grouping, later work in OSN group creation proposed automating group creation while allowing users to modify the created groups [14,16]. Following this philosophy, in 2011, Facebook introduced smart lists.…”

“…One can utilize clustering algorithms to create populated groups from the onset, and then allow for personal curation. While the feasibility of structural network clustering for group creation in social networks has been investigated before [14], less is known about the benefits and drawbacks of using various automated clustering algorithms for grouping people within a social media interface. This work is a first step in that direction.…”