I've worked with the Twitter API before and I really enjoy it, there's a lot of interesting data you can get out of it; in the past I've done projects with natural language processing on tweets, but this time I wanted to do something new.

What it does

You enter a word or phrase, and it pulls from the Twitter API to display geolocation data with Esri. You can add as many queries as you want and see them in superposition on the map together to compare.

How I built it

I originally just used plain HTML/Javascript with an express server, but about halfway through the hackathon I decided to upgrade to Angular so I could improve the UI. I also used a Python backend for communicating with Twitter.

Challenges I ran into

The data is sparser than I'd like, turns out people don't tweet with geolocation turned on very often. From a query of 2000 tweets I'll get less than 20 with a location attached. This can be fixed by just getting more tweets, but that also increases the runtime drastically, so I had to try and find a balance.

Accomplishments that I'm proud of

I really like how it turned out altogether, I'm going to put it up on my website later with my other Twitter-based project that I made at the last hackathon I went to.

What I learned

I learned a lot about the Esri API and I really like it, it was perfect for this project.

What's next for TweetMap

I'll put it up on my website, and maybe spruce up the UI a bit.

Share this project: