Inspiration

We got inspired by the awesome features of Indicoio API and said "why not put it to practice and use it for analyzing contemporary topics?"

What it does

You select two presidential candidates from the top 4 presented. Then you click 'Compare' to see how positive and optimistic their past ~50 tweets have been and also see how their tweets relate to their personality traits. Results are shown using graphs which can be used to compare and analyze trends. Very simple interface.

How We built it

Our website is hosted on Google App Engine and uses webapp2, a python based web-framework. We use twitter API to collect tweets and Indicoio to analyze sentiments and personality.

Challenges we ran into

A rather simple bug in the code that we wrote for graphing the results kept us stuck for 3 hours. JS is not the easiest language to debug.

Accomplishments that we are proud of

We managed to accomplish a simple yet powerful project that makes use of Machine Learning.

What we learned

Machine learning - not as intimidating as it sounds :)

What's next for Positivity in Politics

Make the graphs more interactive and add more details (such as dates and sample tweets) so that users can better understand why the data appears to be the way it is - was it because of the presidential debate that happened few days ago? Does it say anything about the presidential race?

Built With

Share this project:

Updates