The social distributional hypothesis: a pragmatic proxy for homophily in online social networks

  • Folke Mitzlaff
  • Martin Atzmueller
  • Andreas Hotho
  • Gerd Stumme
Original Article

DOI: 10.1007/s13278-014-0216-2

Cite this article as:
Mitzlaff, F., Atzmueller, M., Hotho, A. et al. Soc. Netw. Anal. Min. (2014) 4: 216. doi:10.1007/s13278-014-0216-2
Part of the following topical collections:
  1. Social Systems as Complex Networks

Abstract

Applications of the Social Web are ubiquitous and have become an integral part of everyday life: Users make friends, for example, with the help of online social networks, share thoughts via Twitter, or collaboratively write articles in Wikipedia. All such interactions leave digital traces; thus, users participate in the creation of heterogeneous, distributed, collaborative data collections. In linguistics, the Distributional Hypothesis states that words with similar distributional characteristics tend to be semantically related, i.e., words which occur in similar contexts are assumed to have a similar meaning. Considering users as (social) entities, their distributional characteristics can be observed by collecting interactions in social web applications. Accordingly, we state the social distributional hypothesis: we presume, that users with similar interaction characteristics tend to be related. We conduct a series of experiments on social interaction networks from Twitter, Flickr, and BibSonomy and investigate the relatedness concerning the interactions, their frequency, and the specific interaction characteristics. The results indicate interrelations between structurally similarity of interaction characteristics and semantic relatedness of users, supporting the social distributional hypothesis.

Keywords

Social networks Social interactions Social media  Analysis Distributional semantics  

Copyright information

© Springer-Verlag Wien 2014

Authors and Affiliations

  • Folke Mitzlaff
    • 1
  • Martin Atzmueller
    • 1
  • Andreas Hotho
    • 2
  • Gerd Stumme
    • 1
  1. 1.Knowledge and Data Engineering GroupUniversity of KasselKasselGermany
  2. 2.Data Mining and Information Retrieval GroupUniversity of WuerzburgWuerzburgGermany

Personalised recommendations