Hack CU Twitter LED Map

Team Photo
Map Works!
Close up of map
The map working as planned
Code Receiving Tweets
Initial look of the map
Section of code that pulls tweets from Twitter

Inspiration

We are 4 friends who have majors in the same ballpark (2 electrical engineers, 2 computer science majors), so we were thinking of how we could work together. We knew that Twitter would be at Hack CU IV, so we brainstormed an idea where we could work on hardware and software using the Twitter API.

What it does

Our project reads in tweets using the twee.py library, where it is then evaluated based on the connotation of the language. The evaluation is then represented by an LED map of major Colorado cities. Red represents negative tweets, while green represents positive tweets. White is neutral, pink is neutral-negative, yellow is neutral-positive.

How we built it

We split up into 2 teams, hardware and software. The software team used JSON and the twee.py ftlibrary to acquire the data needed, and honed in on Colorado. The hardware team wired an arduino UNO to a 32x32 LED board, and wrote code translating the evaluation to LED color.

Challenges we ran into

Software Challenges: JSON objects, Using libraries that they never used before, Putting all of the information together, Time delays using the time library

Hardware Challenges: Voltage regulation on the LED board, Having to switch from embedded C to Arduino, Bit shifting, Syncing the software and hardware, Making the system efficient

Accomplishments that we're proud of

Software: Figuring out the JSON objects without using public documentation

Hardware: Figuring out an LED matrix that did not have public documentation

What we learned

Software: JSON, Databases, Time directories, Public API's, Documentation

Hardware: LED matrix coding, Bit math, Bit representation, Arduino, Embedded C,

What's next for Hack CU Twitter LED Map

Hack CU V The entire United States More colors Better gradientsf

Built With

Submitted to

HackCU Episode IV

Created by

I worked on the hardware of the project. It was my job to wire the Arduino up to the LED matrix, so that the information our software guys pulled was able to be visually represented.

Kurt Sprague
I worked on the software to get tweets from the public Streaming API, analyze the language sentiment, and create an averaged happiness score for the top 11 cities in Colorado

lalyon
I was on the software team for this project I wrote the (very simple and crude) NLP that we used to evaluate the sentiment of the tweets. I also worked on understanding the twitter api and creating the stream from twitter and processing the JSON objects the API returned. (I did all this in conjunction with Luke Lyon)

Christian Sousa
Aspiring technology student majoring in computer science at the University of Colorado Boulder. I love cybersecurity and low level computing
Andrew Zhu