Return to Projects

How to detect automaticaly hoaxes ?

With the expansion of social networks, many false or uncertain information (fake) or unverified rumors (hoaxes) are propagated by users. They occur most often by a picture with explanations. The societal impact of these messages is variable (joke, need to fill a lack of image or information about an event) but many within the handling of the deliberate opinion. This project aims to develop new theoretical and technological frameworks to partner with messages (text and images) circulating on the Web (e.g via social networks) with information to check the possible manipulations.

Some examples of fakes/hoaxes :

Our dataset is enable here.

Permanent link to this article:

Facebook pages used

Currently, there are 79 groups in the database and 846 255 messages.

Multimodal approach

Abstract This paper presents a multi-modal hoax detection system composed of text, source, and image analysis. As hoax can be very diverse, we want to analyze several modalities to better detect them. This system is applied in the context of the Verifying Multimedia Use task of MediaEval 2016. Experiments show the performance of each separated …