Broadly speaking, “Fake news” refers to false news stories to deceive or
mislead people. Despite several recent high-profile incidents of fake news (e.g., pizzagate) and wide spread existence of fake news websites, however, we do
not fully understand about fake news yet—i.e., how it is made, who/why makes
them, how it spreads in the network, how it differs from legitimate news, or
why/how people fall for it, etc. The better understanding of fake news would be
able to help people equip better to handle fake news in future. With the risen
interests on the topic and its huge societal implications, therefore, we
believe that this is a right time to embark on a research project on fake news
and make intellectual contributions with broader impacts on society.
In particular, in this SysFake (pronounced as "Cease Fake") project at Penn State, we explore
the possibility of training machines to detect fake news accurately and investigate
whether they can do a better job than humans. We propose to combine complementary perspectives and understanding
on fake news from both computer science and social science disciplines, implement such findings into a
computational model, and carry out validating experiments and user studies.
Our
results will not only have significant intellectual merit in building a
machine-based solution for detecting fake news, but also have far-reaching broader
impact on society and education by helping improve the quality of information
used by citizens to make important decisions in almost all domains.