Cracking Down On Fake News since 2017
Who We Are:
We, Justin Berman, Justin Chiang, and Udai Nagpal, are college students with an interest in business, mathematics, and computer science. We have experience in statistics, machine learning, and computer programming applications from school and outside research projects.
At NewsCracker, our goal is to contribute to the movement against "fake news" by helping everyday Internet users think more critically about the articles they read. We aim to do this by automatically assigning articles a rating and distributing the rating to potential readers of the article before they begin reading. In doing so, we aim to be non-partisan and as objective as possible.
The rating stems from the two main types of "fake news." The first is the reporting of false facts and stories that never happened. Our computer program researches the story being reported while looking at the quotes and references in the article to determine the factual accuracy.
However, few articles of this first type come from widely-read sources, and it's sometimes unfair to penalize articles for factual mistakes, especially with developing stories. Instead, while researching, we noticed that it was much more common for articles to insert opinions into articles labelled as news stories (Of course, there is no issue with publishing opinions under labels such as "Opinion," "Op-Ed," or "Column."). Thus, we are also focusing on using machine learning to detect the second type of "fake news," which we call "spin," or the unnecessary or excessive insertion of opinion into supposedly factual articles.
Overall, our vision is that by raising awareness of potentially misleading articles before they are read, we can play a role in creating a world where news articles are written accurately and objectively, allowing us, as a human race, to make independent and informed decisions.
You may keep scrolling to read our formal mission statement or click here to go into more detail about our methodology.
Our Formal Mission:
NewsCracker utilizes machine learning technology and statistical analysis to determine the accuracy of articles in an objective and non-partisan manner and preemptively attach the label of “fake news” to articles that disingenuously report on their sources in a way that is accessible to a larger, less-informed audience.
Recently, the misreporting of news in general has become a widespread epidemic that has begun to have great sway on the information people use when making social and political decisions. Along with this trend have come advances in large-scale data analysis that enable us to intelligently verify reporting and hopefully curb this phenomenon.
Our vision is to ultimately play a role in contributing to a society in which large-scale, up-to-date, and entirely accurate dissemination of news and opinions is a reality. A society cognizant of the truth is better equipped to make proper decisions for itself and focus on meaningful discussions of more subjective and important issues.
Williams College '21
Brown University '21
Columbia University '21