Inspiration

We were inspired by a common appreciation for horrible puns to annoy our friends. It was as if fate has brought us together to create such an amazing tool to help that one friend become the most annoying in the group.

What It Does

Pun-tionary uses Natural Language Processing to derive context out of an innocent friend's sentence and returns a horrible pun to ruin their day.

How we built it

We used Google's Natural Language Processing API to analyze a user's text and provide context on the sentence. We then grabbed the text and ran it through our special filter which returns a special pun from the Mongo Database from mlab.

We used Python Flask to serve the application and provide endpoints for our React Native application to access. Did we mention that we used Expo XDE to help us build this amazing application blazing fast?!

Challenges we ran into

We definitely had a lot of roadblocks along the way. One of them being our limited understanding of NLP and Machine Learning. It took a while to learn new technologies, but we made it with what we could in such a short time!

What we learned

It's definitely a lot harder to train an ML model to generate puntastic sentences. We realized that in such a short time, we were better off leveraging the NLP tool and using that to derive our own filter to return as-relevant-as-possible puns.

What's next for Pun-tionary?

Open source it! This app has a long way to go, but we believe it'll make it somewhere.

Built With

Share this project:
×

Updates