Title: Clustering user behaviour patterns on Twitter
Authors: Christine Klotz; Coskun Akinalp; Herwig Unger
Addresses: Faculty of Mathematics and Computer Science, FernUniversität in Hagen, Universitätsstraße 27, D-58084 Hagen, Germany ' Faculty of Mathematics and Computer Science, FernUniversität in Hagen, Universitätsstraße 27, D-58084 Hagen, Germany ' Faculty of Mathematics and Computer Science, FernUniversität in Hagen, Universitätsstraße 27, D-58084 Hagen, Germany
Abstract: Personalised systems and targeted services must be tailored to the characteristics of the individual. Several categorisations for Twitter users exist, but all fail to account for the complexity of human beings. Behaviour is a key feature in detecting users with specific characteristics. This study demonstrates how to extract meaningful user behaviour patterns on large-scale datasets that reflect the personalities of human users. This is a first step to prediction of user action and the underlying individual decision-making process.
Keywords: Twitter; tweets; user characteristics; text mining; online social networks; OSN; user behaviour; language; word usage; cluster analysis; k-means clustering; user categorisation; behaviour patterns.
DOI: 10.1504/IJSNM.2016.082632
International Journal of Social Network Mining, 2016 Vol.2 No.3, pp.203 - 223
Accepted: 04 Jun 2015
Published online: 04 Mar 2017 *