Inspiration
We were inspired by the large amount of divisions based on politics in our community
What it does
Fair Reads allows users to submit links to articles to have their biases towards their subject analyzed. Then, Fair Reads selects multiple articles from different sources with different political leans about the same topic for the user to consider as well.
How we built it
We built it using NLTK's natural language processing library to create a machine learning model for sentiment analysis to detect whether or not an article was positive or negative towards a subject and extract its big ideas. Then, we found different news articles with similar topics and used AllSide's database to see the political affiliation of the sources.
Challenges we ran into
We had trouble with preprocessing the data for our model, as we had never done so before.
Accomplishments that we're proud of
We are proud of our simple and elegant UI and accurate model. We are also proud of finding a way to create article recommendations.
What we learned
We learned more about natural language processing.
What's next for Fair Reads
We are going to create a browser add on so our app can be used more quickly.
Log in or sign up for Devpost to join the conversation.