We believe that mental health is important, and today more than ever your mental health can both be influenced and reflected by your social media presense. We wanted to create a simple way for people to be kept up to date with how their mental health is being influenced by social media.
What it does
When someone tweets with the hashtag #TrackMyMood, our program will fetch the recent tweets and analyses them using IBM Watson's tone analyzer to determine whether or not the user is in a bad mental state and offers them the analysis result and possible recommendations on how to get into a better one. The negative emotions will be under the line and positive will be stacked above
How we built it
We used Tweepy to collect Twitter information, then passed that information into IBM Watson's tone tracker API and the plotted the trends using Matpltlib.
Challenges we ran into
The two challenges we ran into were running out of API calls and having trouble creating a graph the properly reflects the data that we want to convey to users. We overcame these opticals by taking our time and doing research on different types of Matpltlib graphs to determine which would most accurately display the information that we wanted to get across.
Accomplishments that we're proud of
Attaching the front end to the back end. Attaching front and back ends is always the biggest obstacle in a hackathon, but we are proud that we were able to seamlessly integrate the front and back end of the project to create a fully functioning project.
What we learned
We learned all about plotting information as well as using IBM and Twitter APIs. (I learned much about python from plotting)
What's next for TrackMyMood
In the future we would like to use the technology we began using to track users information in real time and provide tips to them the moment we notice their mood begin to turn bad. We would also like to create a more visual friendly UI with a web app that displays different types of charts and graphs to give the user a better idea of how their mental state is reflected online.