Interactive Grouping of Friends in OSN: Towards Online Context Management

Bo Gao, Bettina Berendt, Dave Clarke, Ralf De Wolf, Thomas Peetz, Jos Pierson, Rula Sayaf

Research output: Chapter in Book/Report/Conference proceedingConference paper

12 Citations (Scopus)

Abstract

In Online Social Networks (OSNs), it can be
difficult to maintain the context of a conversation or action, i.e.
to know what the situation is and how to act appropriately. The
resulting uncertainties may lead to privacy issues. We focus on
one issue Context Collision in this paper, and motivate that
a first step to address this issue is to help users distinguish
groups of contacts within their OSN accounts. We conducted a
small user study to investigate the criteria of users grouping the
people they know. We summarized our participants strategy of
labeling the groups and found that they perform the grouping
mainly by their connections with others. We used these results
in the design of FreeBu, a semi-automatic and interactive
grouping tool, which is based on mining friend graph data
for community detection and profile information for labeling.
Original languageEnglish
Title of host publicationPinSoDa: Privacy in Social Data - Workshop in conjunction with 12th IEEE International Conference on Data Mining (ICDM 2012), 10 December 2012, Brussels, Belgium, 8
Place of PublicationBrussels
PublisherIEEE Computer Society Press
Pages555-562
Number of pages8
ISBN (Print)978-1-4673-5164-5
Publication statusPublished - 10 Dec 2012
Event2012 IEEE 12th International Conference on Data Mining Workshops - Brussels, Belgium
Duration: 10 Dec 201210 Dec 2012

Conference

Conference2012 IEEE 12th International Conference on Data Mining Workshops
Country/TerritoryBelgium
CityBrussels
Period10/12/1210/12/12

Keywords

  • Online social networks
  • privacy by design
  • Transparancy tool
  • context
  • visualization

Fingerprint

Dive into the research topics of 'Interactive Grouping of Friends in OSN: Towards Online Context Management'. Together they form a unique fingerprint.

Cite this