The science of twitter lists: Understanding membership and subscription through network analysis

Srikar Velichety, Sudha Ram

Research output: Contribution to conferencePaperpeer-review

1 Scopus citations

Abstract

We report on an exploratory analysis of membership and subscription patterns in publicly shared Twitter lists. Our analysis is conducted in two phases. In the first phase, we examine membership and subscription patterns of lists by defining specific quantitative measures. In the second phase, using network analysis, we propose the use of structural holes to assess the assortativity of a network. We examine the partial structure of the network around a node (depicting a list member/subscriber) to discover users' implicit preferences in subscription and membership. We find that distribution of number of list subscribers follows a power law with most lists having very few or no subscribers. We also find that Twitter users usually segregate groups of people and distribute them across their public lists in such a way that there is a very little overlap among the lists. We find similar results in the case of members of lists to which a user has subscribed. We also show that the structure of the network around a list and its curator can help us understand implicit preferences in subscription and membership. Finally, we find that the network characteristics of a list can help predict churn in subscribers more accurately than churn in members.

Original languageEnglish (US)
StatePublished - Jan 1 2013
Externally publishedYes
Event23rd Workshop on Information Technology and Systems: Leveraging Big Data Analytics for Societal Benefits, WITS 2013 - Milan, Italy
Duration: Dec 14 2013Dec 15 2013

Other

Other23rd Workshop on Information Technology and Systems: Leveraging Big Data Analytics for Societal Benefits, WITS 2013
Country/TerritoryItaly
CityMilan
Period12/14/1312/15/13

ASJC Scopus subject areas

  • Information Systems

Fingerprint

Dive into the research topics of 'The science of twitter lists: Understanding membership and subscription through network analysis'. Together they form a unique fingerprint.

Cite this