Abstract |
Over the past few years, we have been witnessing the rise of misinformation on the Internet.
People fall victims of fake news continuously and contribute to their propagation knowingly or
inadvertently. The use of propaganda is indeed ancient, but never before has there been the
technology to so effectively disseminate it. The social media engagement that has swept our lives
over the past decade practically exploded the proliferation of misinformation, including the
associated distribution of fake news. The ‘pizzagate shooting’ incident and the Cambridge
Analytica scandal indicate that we should not take this rise of misinformation lightly.
Many recent efforts seek to reduce the damage caused by fake news by identifying them
automatically with artificial intelligence techniques, using signals from domain flag-lists, online
social networks, etc.
In this thesis, we present Check-It, a system that combines a variety of signals into a pipeline for
fake news identification. Such signals include the reputation of the person (account) posting the
news, the reputation of the website where the news is hosted, the linguistic features that
characterize a fake news article as well as the article’s content per se.
Using a deep learning approach, we combine all these features towards providing a rating that is
timely and accurate. Check-It is developed as a web browser plugin with the objective of efficient
and timely fake news detection while respecting user privacy. The requirements we considered
when designing Check-It is GDPR compliant, highly confident identification, low response time
and lightweight computation. To implement our plugin, we have used pure JavaScript
frameworks, like Minhash.js and TensorFlow.js.
In this thesis, we present the design, implementation, and performance evaluation of Check-It.
Experimental results show that it outperforms state-of-the-art methods on commonly used
datasets while achieving an accuracy of 93%. Furthermore, we provide some directions that can
guide future versions of Check-It.
|