Inspiration
We wanted to explore the data from a public health lens, looking to determine whether there are certain variables related to hospital readmittance.
What it does
We built a multinomial logistic regression model that predicts subject probability of being readmitted versus not being readmitted.
How we built it
Preliminary data analysis was conducted in Tableau and data cleaning and modeling was done in RStudio
Challenges we ran into
Identifying motivating question/goals and validating the model.
Accomplishments that we're proud of
Constructing a working predictive model.
What we learned
How to validate multinomial logistic regression model, fundamental logistic regression model concepts, and utilizing Tableau for data visualization
What's next for Predicting Readmittance for Diabetes Patients - CDC 2023
Combining predictor variables together to reduce model complexity, addressing skewed data issue, using different predictive model functions like decision trees to improve accuracy of predictions.
Log in or sign up for Devpost to join the conversation.