Googling things is a central part of being any student. To quickly evaluate the contents of a source is crucial when cramming for projects, essays and the like. This extension allows you to take a tl;dr (Too long; Didn't Read) before even clicking on a link giving you valuable insight while saving time. Our greatest inspiration comes from the one and only Front Page of the Internet - Reddit's Enhancement Suite.

What it does

We have created an innovative Chrome Plugin which uses Watson's NLP Analysis and Natural Language Toolkit to create a summary of webpages at the click of a button.

How we built it

Our program is a cloud hosted Python back-end server powered by Heroku. Client side, we used Chrome's Extension Libraries to create a product that is simple to install and use. Watson was used to identify key words and Python's Natural Language Toolkit along with Google's tensorflow to filter, deconstruct and reconstruct the data into something readable and effective summary

Challenges we ran into

Heroku nltk dependencies are buggy and hacky to incorporate.

Accomplishments that we're proud of

Effective summation of most articles, especially with respect to Wikipedia.

What we learned

Languages are difficult to understand using software, and the integration of multiple systems together is also more complex than it appears

What's next for Chrome TLDR

Multiple and auto language support and detection. Optimization for specific web sites ie Amazon. PDF Support for research papers, image recognition for relevant images.

Share this project: