Article: A study of different keyword activity prediction problems in social media Journal: International Journal of Social Network Mining (IJSNM) 2016 Vol.2 No.3 pp.224 - 255 Abstract: Forecasting keyword activities in social networking sites has been the subject of many studies, as such activities represent, in many cases, a direct estimate of the spread of real-world phenomena, e.g. box-office revenues or flu epidemic. Most of these studies rely on point-wise, regression-like prediction algorithms and focus on few, usually unambiguous, keywords. We study in this paper the impact of keyword activity on three different problems: a) classification of keywords according to the increase of their activity in the near future; b) prediction of the activity value of each keyword in the near future; c) ranking of a set of keywords according to their future activity values. It is the first time, to our knowledge, that such dimensions are evaluated in this framework. Our experiments are conducted on a large dataset built by monitoring Twitter over a year. The different methods tested are evaluated using standard scores as well as a newly defined, application driven quality measure. Inderscience Publishers - linking academia, business and industry through research

Title: A study of different keyword activity prediction problems in social media

Authors: François Kawala; Éric Gaussier; Ahlame Douzal-Chouakria; Eustache Diemert

Addresses: Université Joseph Fourrier, Grenoble 1, CNRS, Laboratoire LIG, Équipe AMA, Bat. CE4, allée de la palestine, F-38610 Gières, France ' Université Joseph Fourrier, Grenoble 1, CNRS, Laboratoire LIG, Équipe AMA, Bat. CE4, allée de la palestine, F-38610 Gières, France ' Université Joseph Fourrier, Grenoble 1, CNRS, Laboratoire LIG, Équipe AMA, Bat. CE4, allée de la palestine, F-38610 Gières, France ' Société Purch, 4 rue des méridiens Parc Sud Galaxie, F-38130 Echirolles, France

Abstract: Forecasting keyword activities in social networking sites has been the subject of many studies, as such activities represent, in many cases, a direct estimate of the spread of real-world phenomena, e.g. box-office revenues or flu epidemic. Most of these studies rely on point-wise, regression-like prediction algorithms and focus on few, usually unambiguous, keywords. We study in this paper the impact of keyword activity on three different problems: a) classification of keywords according to the increase of their activity in the near future; b) prediction of the activity value of each keyword in the near future; c) ranking of a set of keywords according to their future activity values. It is the first time, to our knowledge, that such dimensions are evaluated in this framework. Our experiments are conducted on a large dataset built by monitoring Twitter over a year. The different methods tested are evaluated using standard scores as well as a newly defined, application driven quality measure.

Keywords: web mining; clustering; classification; association rules mining; empirical studies; keyword activity prediction; social media; social networking sites SNS; Twitter; tweets.

DOI: 10.1504/IJSNM.2016.082642

International Journal of Social Network Mining, 2016 Vol.2 No.3, pp.224 - 255

Accepted: 08 Oct 2015
Published online: 04 Mar 2017 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article

Title: A study of different keyword activity prediction problems in social media

Keep up-to-date