Inspiration
As high school and college students, we understand the common practice of putting on a smile for your friends so they don’t worry about you or venting common difficulties that all your friends can relate to and laugh about. However, social media is also a platform on which we can become our most vulnerable selves.
What it does
Our product has three main features.
Our sentiment analysis feature uses a python server and IBM Watson Natural Language Processing to filter in tweets and gives an overall score for depression. Our frontend will give a rating telling the friend is either "normal" or "depressed".
Our Community Matchmaker imports data from Google Firebase. When users signup, they indicate if they want to be a mentor or mentee and their profile information will appear on our Community Page along with a button to contact them.
Our Resources Page gives helpful links for those who believe they need professional help.
How we built it
We used ReactJS to build the frontend and python for the backend and twitter sentiment
Challenges we ran into
Our main trouble was connecting the frontend and backend, specifically displaying the JSON data from the twitter sentiment analysis onto our website.
Another problem is to develop a reliable way to classify depressed tweets and normal ones
Accomplishments that we're proud of
We are proud of not only creating a feature using machine learning, but providing a platform for communities to grow amongst other sharing the same struggle.
What we learned
We learned that full stack development is a very useful tool and can allow for greater versatility when making products.
What's next for HealthComm
One development is to further train our sentiment analyzer to detect more keywords and determine sentiment such as positivity, calm, stress, etc. rather than just depression which is a big assumption to make. Another further development is to create groups that users can join.
Log in or sign up for Devpost to join the conversation.