This is an archive for all notebook files created for the DS4A Project (Team 54).
Effective 2021, the federal government promulgated a new rule for hospitals requiring them to disclose pricing on a wide range of hospital services. This project explored how this price transparency regulation can help patients understand costs, and examine the relationship between hospital pricing and demographic data in Michigan.
See the Project Feature. Or read the Final Report of this project.
-
data folder
- df_hospital_clean.csv (hospital data)
- df_population.csv (demographic data)
- df_price_clean.csv (pricing data)
-
data_schema_diagram.drawio
This is the diagram for the data schema. -
batch_processing_excel_files.ipynb
This notebook is created to batch process excel files. It reads and extracts a selection of features by cpt_code. -
dataset_merging_check.ipynb
This notebook prepares all dataframes for the merging. It checks foregin keys for typos, erros, and potentially missing values. -
eda_nyembo.ipynb
This notebook is the first version of eda, developed by kasole nyembo. -
knn_imputation.ipynb
This notebook is the first implementation of knn imputation, which did not end up in the final product. -
regression_model.ipynb
This notebook is the main file for predictive modeling. In addition to model selection/optimization, it also contains correlation matrix and pca biplot for the eda.