This project analyzes a dataset of 9,105 critically ill patients from various U.S. medical centers. Our aim is to use machine learning to improve end-of-life care decisions.
The dataset, sourced from UCI Machine Learning Repository, encompasses a diverse range of medical and demographic variables. It provides a comprehensive view of patient care in critical conditions.
- To conduct a full-scale analysis using machine learning methods taught in class.
- Utilize models like Linear Models, Generalized Linear Models, Generalized Additive Model, and others.
- Focus on delivering insights that could be pivotal in critical healthcare decision-making.
- Data Preprocessing: Address missing values, encode categorical data, and perform feature engineering.
- Exploratory Data Analysis (EDA): Examine data distributions, correlations, and key patterns.
- Model Development: Apply various machine learning techniques.
- Model Evaluation: Use metrics like RMSE for model comparison.
- Interpretation: Provide a comprehensive interpretation of the findings and their implications.
- R Language: Main programming language used.
- R Libraries: Including
caret
,naniar
,ggplot2
, etc.
...
- Samuel Devdas
- Pradip Ravichandran
Thanks to the providers of the dataset and our course instructors for guidance.
'All Rights Reserved'
For inquiries, please contact
Note: This analysis is part of our academic project in the Machine Learning course. The findings and methodologies are subject to academic scrutiny.