TiDeH: Time-dependent Hawkes process for predicting retweet dynamics

Ryota Kobayashi, Renaud Lambiotte

Research output: Contribution in Book/Catalog/Report/Conference proceedingConference contribution

Abstract

Online social networking services allow their users to post content in the form of text, images or videos. The main mechanism driving content diffusion is the possibility for users to re-share the content posted by their social connections, which may then cascade across the system. A fundamental problem when studying information cascades is the possibility to develop sound mathematical models, whose parameters can be calibrated on empirical data, in order to predict the future course of a cascade after a window of observation. In this paper, we focus on Twitter and, in particular, on the temporal patterns of retweet activity for an original tweet. We model the system by Time-Dependent Hawkes process (TiDeH), which properly takes into account the circadian nature of the users and the aging of information. The input of the prediction model are observed retweet times and structural information about the underlying social network. We develop a procedure for parameter optimization and for predicting the future profiles of retweet activity at different time resolutions. We validate our methodology on a large corpus of Twitter data and demonstrate its systematic improvement over existing approaches in all the time regimes.

Original languageEnglish
Title of host publicationProceedings of the 10th International Conference on Web and Social Media, ICWSM 2016
PublisherAAAI Press
Pages191-200
Number of pages10
ISBN (Electronic)9781577357582
Publication statusPublished - 2016
Event10th International Conference on Web and Social Media, ICWSM 2016 - Cologne, Germany
Duration: 17 May 201620 May 2016

Conference

Conference10th International Conference on Web and Social Media, ICWSM 2016
CountryGermany
CityCologne
Period17/05/1620/05/16

Fingerprint Dive into the research topics of 'TiDeH: Time-dependent Hawkes process for predicting retweet dynamics'. Together they form a unique fingerprint.

Cite this