GottaGo

Unlock the power of AI for personalized mental well-being. Our AI tailors low-effort events to boost happiness in SF, connecting users to personalized meaningful experiences in the city.

Comment

Inspiration:

San Francisco is a city of contrasts, with both vibrant happiness and looming mental well-being challenges. Over 51% of its residents battle mood and stress disorders, and traditional therapies and medications are often out of reach. However, we believe in the power of everyday actions to improve mental well-being: cooking, dancing, physical activity, volunteering, socializing, and enjoying personal interests. Unfortunately, when you're in a tough mental state, you're not in a position to find these experiences. Introducing GottaGo: a low-effort, personalized recommender in a Tinder-style format.

What It Does:

GottaGo recommends curated activities that enhance mental well-being. Swipe through personalized suggestions and discover experiences that can lift your spirits and improve your mood.

How we built it

  1. User research: to identify a moment with an unmet need for our target market of SF community members.
  2. Research for datasets: meetup groups, event link websites, facebook communities, mental health socials
  3. Downloaded the event descriptions and generated embeddings for the dataset. Used the generated embeddings to look for similarities and cluster similar activities and events
  4. Wrote the code for the recommendation algorithm, the algorithm depends on user feedback to give the next recommendation. If the user dislikes the current recommendation the algorithm learns that behaviors' and recommends activities from a different cluster. If the user likes the present recommendation the recommender recommends similar activities from the same cluster.
  5. Used Dall-E to generate images for each of the events based on the event decsription
  6. Integrated the python based recommender algorithm with the javascript UI .

Challenges we ran into

UI design : None of our team members had previous experience in UI design and thus there is a strong need to improve the UI experience of our users when we deploy this.

Data parsing and cleaning the dataset: Sometime our parsed data would find similarities with other unrelated events based on specific words like the month or the year. We would need to clean the dataset such that it only finds meaningful similarities.

Accomplishments that we're proud of

Used AI at each and every stage. The Demo actually works!

What we learned

71% of tech workers said their productivity is affected by a mental health issue 57% of tech industry employees reported burnout Importance of proactively looking after mental wellbeing in day to day life rather than treating the mood later.

What's next for GottaGo

Do things at scale , that scale. Update the data daily in real-time, engage with the event organizers to use our platform to showcase their events. Do more customer research and improve the UI based on their feedback. Add automations to clean and parse the data.

Built With

Share this project:

Updates