Our team wanted to build a tool to help track the warning signs of depression in people. We often have a bad perception of the impact of our own words and with Mindful, you can track the negativity of your words over time.
What it does
How we built it
We began with writing the backend as we had assumed it would be the more time-consuming part of the project. We used Node and the Twitter API to get a user's tweets and photos that they've tweeted. We send these images through Microsoft's Project Oxford facial recognition API and scan through the photos looking for the user and then analyze the facial expressions present in the picture. The text tweets are sent through IBM's Watson platform and the text is analyzed looking for emotional categorization. The data is aggregated and presented to the user on the final page where they can see information about the analysis including a graph of the timeline of emotions.
Challenges we ran into
Accomplishments that we're proud of
What we learned
What's next for Mindful