Reports

May 17, 2018  |  Updated June 4, 2018

Developing a Disinformation Detection System and Sourcing it Live: The Case Study of the 2018 Italian Elections

By Alexandre Alaphilippe

During the past months, the phenomenon of fake news has drawn significant attention from both researchers and civil society actors. Following the promising results obtained during the 2017 French elections in the identification and sourcing of disinformation content, we tested the implementation of a monitoring system based on an identified ecosystem rather than on a series of keywords.
The purpose of this monitoring process was to give the ability to expose misinformation networks in real time.

As a testing ground for our method we chose the 2018 Italian General Election. During our monitoring we’ve received 171 alerts from the system, with the content recognized as fake, misleading or manipulated a total of 54 times. The Twitter exposure of our findings managed to gather 109,000 impressions.

View PDF