This project is part of the ADS-503 course in the Applied Data Science Program at the University of San Diego, taught by Professor Satya Varaprasad Allumallu.
- Data Mining
- Predictive Modeling
- Machine Learning
- Data Visualization
- Data Engineering
- R
This analysis aims to predict inpatient hospital readmissions within 30 days among diabetic patients. In 2019, diabetes affected an estimated 37.3 million individuals in the United States; forecasting readmission can elicit critical insights for medical professionals. This analysis uses a dataset provided from Kaggle, containing information from diabetic hospital encounters spanning 1999-2008. The dataset contains 101,766 observations and 50 columns; 49 predictor features, and one target feature. The target feature is a categorical variable that specifies whether a patient was “not readmitted,” “readmitted within 30 days,” or “readmitted after 30 days.” However, we converted this into a binary variable classifying whether a patient was readmitted within 30 days or not.
See data dictionary here: https://archive.ics.uci.edu/dataset/296/diabetes+130-us+hospitals+for+years+1999-2008
R version 4.1.2 (2021-11-01) -- "Bird Hippie" Copyright (C) 2021 The R Foundation for Statistical Computing Platform: x86_64-apple-darwin17.0 (64-bit)