Reddit and other forum bots that create summaries of posts or articles linked by users of the site. Currently studying Natural Language Engineering at Uni and wanted to implement something using this.

What it does

Creates a 3 sentence summary for any topic inputted by the user. For example, the user could search for something they want to learn about and receive a brief paragraph describing topic concisely. TL;DR in motion - gif

How I built it

Using NLE I created a web app using HTML, CSS and Javascript. Called the 'SMMRY' API to find info for a topic entered by the user before shortening it to 3 sentences through the use of NLE.

Challenges I ran into

I had never used an external API before in a project and the SMMRY API is poorly documented so I had to learn how to run GET requests using APIs in javascript.

Accomplishments that I'm proud of

Having a working webapp at the end of the hack using APIs

What I learned

JQuery API calls,

What's next for TL;DR-ify

To improve TL;DR-ify I would need to improve the method of summarising the topic inputted by the user. I would do this by searching more sites for the information to turn into the TL;DR. The natural engineering criteria determining what is important for the TL;DR could also be improved. Finally, if I wanted to extend it beyond an interesting website, I could attempt to use the base algorithms to create a forum bot similar to the ones that inspired my project to come full circle.

Built With

Share this project: