Project: Library Team 3

Team Members:

Anya Raspopovic, Kevin Marko, Candace Sheremeta, Alex Hopkins, Ricky Hopper, and Crystal Martin

Tagline:

Words of NC

Problem: 

Hunt library has several large screens that they want to have dynamic visuals on. We made a map of North Carolina using universities as points of interest to create word clouds which are gathering the hashtags from twitter.

How it works:

This app first uses Twitter's REST API to pull the most recent 60 tweets from 10 certain geographic areas around the state of North Carolina. Each of these locations represents a 5-mile radius around a university in NC. The app filters the results of the texts in these tweets and passes the filtered text to a function that builds word clouds. The word cloud takes into account how often each word was seen and places the most used words in the word cloud, basing the word's size in the cloud on the number of times the word was seen. The word cloud also takes each university's colors in as parameters and only uses those colors in each respective cloud. The app then displays these word clouds on a map of North Carolina, located in the approximate position of the university that each word cloud represents. The update times for each word cloud are staggered so that, while each individual word cloud only updates every 20 seconds, the app updates the whole map every 2 seconds. Ultimately, this works to give a real-time perspective to what students around the state are discussing on Twitter.

Unfinished and future work:

There are a couple of changes that we would like to make to this prototype in the future. Of primary concern is making the app itself more aesthetically appealing by changing the word clouds from a square/rectangle to a circle. Unfortunately, there was no support for this feature in the word cloud generator we used as of the time of this prototype. It would also be nice if we could make the size of the words more dynamic for each cloud. As of now, we are using a constant to determine the size of each word in the word clouds. However, if we were able to change the size of the words based on the relative number of times the word was seen based on the other words in the cloud (instead of the words in the entire map), it may be more visual appealing and informative.

Links:

Working Site
Video
Github