Inspiration
Our inspiration came from seeing eye-catching statistics on both the prevalence and treatment, or lack thereof, of mental health among the general population. We thought that we could provide some support to those who don't have access to expensive mental health services, such as low income individuals and underprivileged communities.
What it does
Our app takes in user information and predicts what mental health condition the user could be experiencing. It also sends the user's question to a mental health specialist to answer the question.
How we built it
We started by ideating a problem that could be solved using the tools we had available to us. We began by looking for different services that could benefit our concept of providing realtime information to users in addition to sending information to other sources. We decided on using CockroachDB, Cohere AI, in addition to supporting those services using CSS, HTML, Python3, and SQL.
Challenges we ran into
Some of the challenges we ran into were integrating all of the different components we were using, especially with services that we were not familiar with before the hackathon. CockroachDB gave us some issues trying to understand the API and utilizing it into a local application.
Accomplishments that we're proud of
We are proud of incorporating CockroachDB and Cohere AI's API's into our project and building a website which has a positive impact on the health and safety of our users.
What we learned
We learned how to efficiently push and get information from the CockroachDB database service , and also how to use the NLP toolkit available on Cohere AI. We learnt how to use embedding to classify words and compute the cosine similarity of the inputted words in comparison to our pre-existing classification.
What's next for Mellow
Mellow's next steps will be to expand our network of mental health specialists and expand our consideration of more mental health conditions to continue to provide accurate information to users.
Log in or sign up for Devpost to join the conversation.