Projecte R+D+i amb finançament extern
Project PCI2020-120689-2 DISSIMILAR
Detection of fake newS on SocIal MedIa pLAtfoRms
Of National scope.
This project combines research on watermarking and machine learning with a user experience study to develop novel technological tools to help users to distinguish between original and altered media content. With the proposed tools, we hypothesize that online social media users will be able to identify the authorship of a content to distinguish legitimate from fake multimedia content in an autonomous manner, without the need for the platform manager to control or validate any content. Furthermore, the watermarking tools will also provide content creators with a way to protect their creations against manipulation. Hence, the aim of this project is to provide user-centric tools to face disinformation and that contribute minimizing the redistribution of fake news in online social media. Nowadays, digital watermarking is recognized as a promising technique developed to address the problems of copyright protection, content authentication, tamper detection, and others. Digital watermarking for fake news identification and tracing is novel and interesting. Regarding detection tools, there have been some attempts to counter deep fakes or fake news in images, however they have not been proposed and analyzed for other types of digital content and they have never been applied in a more complete system integrated with social media platforms. Therefore, we can conclude that the proposed approach has innovative potential. Specific objectives: This project can be divided in two main...
Researchers
Former participants (4)
- Blanche Tarragó, Daniel 20212022
- Morey Cortès, Francisca 20232023
- Peiruza Parga, Joel 20232024
- Suárez Gonzalo, Sara Paz 20242024