Skip to Main content Skip to Navigation
New interface
Conference papers

Collaborative Content-Based Method for Estimating User Reputation in Online Forums

Amine Abdaoui 1 Jérôme Azé 1 Sandra Bringay 1, 2 Pascal Poncelet 1 
1 ADVANSE - ADVanced Analytics for data SciencE
LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier
Abstract : Collaborative ratings of forum posts have been successfully applied in order to infer the reputations of forum users. Famous websites such as Slashdot or Stack Exchange allow their users to score messages in order to evaluate their content. These scores can be aggregated for each user in order to compute a reputation value in the forum. However, explicit rating functionalities are rarely used in many online communities such as health forums. At the same time, the textual content of the messages can reveal a lot of information regarding the trust that users have in the posted information. In this work, we propose to use these hidden expressions of trust in order to estimate user reputation in online forums.
Complete list of metadata

Cited literature [9 references]  Display  Hide  Download
Contributor : Amine Abdaoui Connect in order to contact the contributor
Submitted on : Saturday, October 31, 2015 - 3:49:44 PM
Last modification on : Friday, August 5, 2022 - 3:02:49 PM
Long-term archiving on: : Friday, April 28, 2017 - 6:30:43 AM


Files produced by the author(s)



Amine Abdaoui, Jérôme Azé, Sandra Bringay, Pascal Poncelet. Collaborative Content-Based Method for Estimating User Reputation in Online Forums. WISE: Web Information Systems Engineering, Nov 2015, Miami, United States. pp.292-299, ⟨10.1007/978-3-319-26187-4_26⟩. ⟨lirmm-01222961⟩



Record views


Files downloads