Slippery News Detection

logo
extension
flagged content

Inspiration

We hate people selling us lies, so we decided to call them out on their bullshit. With our extension, anyone can find out what news are worthy of their time and who is only "clickbaiting" them. Let's put an end to media exaggeration, sensational headlines and overall low quality of the content just by clicking a button.

What it does

It takes the article in question through multiple APIs to categorise its content. We check the authenticity of images, bias of the title and article itself and how trustworthy the domain owner is. Based on all that, we generate its overall score with an algorithm and give feedback to users in form of visualisation. For those more curious, we also offer some additional information about the article on a separate web platform. Users can even notify us about any web page they believe might be fake or have misleading information, which we also take into account in our bias calculations.

How we built it

We used Google Chrome Extension functionality to manipulate the pages, APIs to scrape and evaluate the content and some of our own algorithms as well. We built a database of scored articles and web application for additional visualisations, which we have deployed on a Heroku server.

Challenges we ran into

Google Chrome Extension has limited functionality, so we had to work around it sometimes :P Also, just the lack of time since we were learning on the go.

Accomplishments that we're proud of

Finishing strong, being awesome at teamwork and still managing to deliver a working prototype.

What we learned

We learned how to overcome programming challenges without becoming crazy, how to connect different technologies and manipulate webpages and much more! :)

What's next for Slippery News Detection

Much is still to be done. We plan to perfect and monetise the idea and improve our machine learning algorithms with gathering more data and seeking out professional help.