Home // ICDS 2014, The Eighth International Conference on Digital Society // View article

Detection of Persuasion Campaigns on Twitter™ by SAX-VSM Technology

Authors:
Sergey Malinchik

Keywords: Twitter™; advertisement; persuasion detection

Abstract:
In this paper, we present a novel approach for detection of persuasion campaigns in online social networks. We demonstrate that temporal evolution of different information cascades in social media display unique signatures of diffusion patterns which are indicators of different kinds of information spreads in underlying networks. We describe a progress of information diffusion through networks by multidimensional time series, representing temporal behavior of multiple cascade features and apply our SAX-VSM technique for classification of the time series. This approach allows us to distinguish two types of topics on Twitter™, promoted or advertisement campaigns and non-promoted or naturally trending topics. We show that the classification can be done without content analysis of topics, using only network topological features, statistics of users’ temporal activity within networks, and some metadata. Optimal selection of right information cascade features allows to achieve classification accuracy ~ 97%.

Pages: 117 to 122

Copyright: Copyright (c) IARIA, 2014

Publication date: March 23, 2014

Published in: conference

ISSN: 2308-3956

ISBN: 978-1-61208-324-7

Location: Barcelona, Spain

Dates: from March 23, 2014 to March 27, 2014