Inspiration
To predict future cases of dementia in order to facilitate the proper allocation of funds
What it does
Use drug prescriptions as a proxy to predict dementia incidence
How we built it
- Data wrangling
- Linear regression
- Long short-term memory (LSTM)
Challenges we ran into
- Data availability
- Data cleaning
Accomplishments that we're proud of
We managed to generate a time-series dataset that could easily be used with different machine learning models to predict dementia incidence
What we learned
- Teamwork is key!
- Time management
- Data is not always clean
Built With
- keras
- python
- r
- scikit-learn
Log in or sign up for Devpost to join the conversation.