Inspiration
When looking through the news in the modern decade, one can find it hard to discern objectivity and bias. It can be very tedious to fact-check an article by yourself, especially in this day and age with thousands of articles being posted every day. A prime example of this is on Reddit, where any user is easily able to post news articles to a subreddit for the whole Internet to see. Observing this made us think, “Why not create a resource that compiles all the information in one place that would allow the user to draw their own conclusions on a Reddit article smoothly?”
What it does
Validdit shows the user useful information about a Reddit post, such as the upvote/downvote ratio, info/history about the poster, comments with links to other news articles, and comments that contain keywords that may question or help verify the validity of the post.
How we built it
HTML, CSS, and JS were used to build the frontend. Python was used to build the backend, working in tandem with the Frame web framework to communicate with the frontend. PRAW, also known as the Python Reddit API Wrapper, was used to access the Reddit API and receive public information from the site, such as a post's information and comments.
Challenges we ran into
- Linking backend to frontend was something we had to learn on the fly and adapt to the new syntax
- Using and tuning the Reddit API in favour of what we were trying to achieve
- Making a fully responsive and aesthetic webpage
- Had to learn how to use Jinja 2 syntax and conditionals along with range statements in order to send list objects from the backend to the frontend
Accomplishments that we're proud of
- Making a functional backend from an API that would get the data we desired (using PRAW)
- Creating a fully functional and working website with great aesthetics (fully responsive)
- Learning new syntaxes and how to use backend and frontend code in tangent
What we learned
We learned how to use PRAW (Python Reddit API Wrapper) in order to obtain public information from the Reddit website such as post information and comments, which are essential pieces of information that are needed for Validdit to work. We also learned how to use the Flask Python framework and Jinja, a web app framework and a web template engine that helped us create the Validdit website and have it communicate with our Python code.
What's next for Validdit
For the future, Validdit can branch-off and expand into other web forums besides Reddit. Machine learning can be implemented to help find keywords that lead to more controversial and interesting comments that help the reader see an even better picture of a news article.

Log in or sign up for Devpost to join the conversation.