Inspiration

So many people around the world, including those dear to us, suffer from mental health issues such as depression. Here in Berkeley, for example, the resources put aside to combat these problems are constrained. Journaling is one method commonly employed to fight mental issues; it evokes mindfulness and provides greater sense of confidence and self-identity.

What it does

SmartJournal is a place for people to write entries into an online journal. These entries are then routed to and monitored by a therapist, who can see the journals of multiple people under their care. The entries are analyzed via Natural Language Processing and data analytics to give the therapist better information with which they can help their patient, such as an evolving sentiment and scans for problematic language. The therapist in turn monitors these journals with the help of these statistics and can give feedback to their patients.

How we built it

We built the web application using the Flask web framework, with Firebase acting as our backend. Additionally, we utilized Microsoft Azure for sentiment analysis and Key Phrase Extraction. We linked everything together using HTML, CSS, and Native Javascript.

Challenges we ran into

We struggled with vectorizing lots of Tweets to figure out key phrases linked with depression, and it was very hard to test as every time we did so we would have to wait another 40 minutes. However, it ended up working out finally in the end!

Accomplishments that we're proud of

We managed to navigate through Microsoft Azure and implement Firebase correctly. It was really cool building a live application over the course of this hackathon and we are happy that we were able to tie everything together at the end, even if at times it seemed very difficult

What we learned

We learned a lot about Natural Language Processing, both naively doing analysis and utilizing other resources. Additionally, we gained a lot of web development experience from trial and error.

What's next for SmartJournal

We aim to provide better analysis on the actual journal entires to further aid the therapist in their treatments, and moreover to potentially actually launch the web application as we feel that it could be really useful for a lot of people in our community.

Share this project:

Updates