The Fake News Challenge aims at combating the problem of fake news using AI/Machine Learning. As a first step towards that goal, it focuses on the problem of Stance Detection, wherein we aim to automcatically assess the stance of the body of a news article with respect to it’s title. The rationale behind this is that if multiple credible news sources show a positive stance towards the same (or similar) titles, then the news is likely to be genuine. On the other hand, if sources with less credibility show a positive stance, the credibility of the news is deemed to be low as well.
We use traditional feature engineering techniques for extracting a variety of relevant signals from the text. Our team (OSUfnc) ranked 7th of the 50 teams that participated