Extracting and Measuring Relationship Strength in Social Networks

View Sample PDF

Author(s): Steven Gustafson (GE Global Research, USA)and Abha Moitra (GE Global Research, USA)
Copyright: 2012
Pages: 15
Source title: Social Networking and Community Behavior Modeling: Qualitative and Quantitative Measures
Source Author(s)/Editor(s): Maytham Safar (Kuwait University, Kuwait)and Khaled Mahdi (Kuwait University, Kuwait)
DOI: 10.4018/978-1-61350-444-4.ch010

Keywords: Information Science Reference / Media & Communications / Social Computing / Social Networking

Purchase

View on the publisher's website for pricing and purchasing information.

Abstract

This study examines how extracting relationships from data can lead to very different social networks. The chapter uses online message board data to define a relationship between two authors. After applying a threshold on the number of communications between members, the authors further constrain relationships to be supported by each member in the relationship also having a relationship to the same third member: the triangle constraint. By increasing the number of communications required to have a valid relationships between members, they see very different social networks being constructed. Authors find that the subtle design choices that are made when extracting relationships can lead to different networks, and that the variation itself could be useful for classifying and segmenting nodes in the network. For example, if a node is ‘central’ across different approaches to extracting relationships, one could assume with more confidence that the node is indeed ‘central’. Lastly, the chapter studies how future communication occurs between members and their ego-networks from prior data. By increasing the communication requirements to extract valid relationships, it is seen how future communication prediction is impacted and how social network design choices could be better informed by understanding these variations.

The IRMA Community

Research IRM

Extracting and Measuring Relationship Strength in Social Networks

Purchase

Abstract

Related Content

IRMA Sponsors