Friends are forever? Evolution of active friendship clusters in online social networks

Haritabh Gupta, Shamik Sural, Niloy Ganguly, Jaideep Vaidya

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

Abstract

The goal of this paper is to study how friendship clusters evolve in online social networks. Results obtained from our work on Facebook data indicate that the set of friends who actively interact during a particular time interval is only a fraction of the total number of listed friends and this set of active friends tends to evolve with time. Interaction footprints on the timeline are used to cluster such active friends. Since there is a temporal variation in these clusters, it is modeled as a graph, wherein nodes represent cluster memberships and edge weights represent the number of common members between a pair of nodes. Novel methodologies for predicting the size and membership of the next active cluster based on the features extracted from past interaction of a user are introduced and their performance evaluated.

Original languageEnglish (US)
Title of host publication2017 9th International Conference on Communication Systems and Networks, COMSNETS 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages558-563
Number of pages6
ISBN (Electronic)9781509042500
DOIs
StatePublished - Jun 9 2017
Event9th International Conference on Communication Systems and Networks, COMSNETS 2017 - Bangalore, India
Duration: Jan 4 2017Jan 8 2017

Publication series

Name2017 9th International Conference on Communication Systems and Networks, COMSNETS 2017

Other

Other9th International Conference on Communication Systems and Networks, COMSNETS 2017
Country/TerritoryIndia
CityBangalore
Period1/4/171/8/17

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications
  • Computer Science Applications

Keywords

  • Cluster evolution
  • Friendship cluster
  • Social network

Fingerprint

Dive into the research topics of 'Friends are forever? Evolution of active friendship clusters in online social networks'. Together they form a unique fingerprint.

Cite this